![[HaBo]](/styles/default/logoklein.png){kind=small}
Mackz
Member of Honour
Thank you for joining today?s chat about Longhorn / Windows Vista Speech Recognition Team
Q: [9] Is a noise cancelling mic necessary for good results?
A: A good microphone will definitively improve the quality of recognition, especially in dictation. Noise cancelling microphones will definitively boost your accuracy in a moderate noisy environments.
Q: [6] will there be any conflicts with running Dragon Natural, or is there a gap between the two that would even require the need for it
A: There shouldn't be any conflicts between Dragon Naturally Speaking and Windows Vista Speech Recognition. You should be able to run either one on Windows Vista.
Q: [2] speech recognition will be on spanish language ?
A: Spanish is one of the languages we are actively working on for Vista. We are simultaneously working on several languages, but we haven't made our final decisions on which languages will be shipped with Vista.
Q: [3] Does the french recognition coming ?
A: French is also a language we are actively working on for Vista
Q: [12] I've been afraid to use voice for a number of reasons. 1. it doesn't really work very well and 2. it really uses a lot of computer resources and can cause a lot of problems. Is this still true?
A: We've tried really hard in Windows Vista to make it easier to use and to be much more accurate. We hope that you've seen some of that in the CTPs and in Beta 1.
Q: [7] Is Polish recognition coming to Windows some day?
A: We don't have Polish scheduled in the short term. Thank you for that suggestion!
Q: [8] For how many languages the Microsoft intends to launch the system of voice recognition?
A: We are actively working on several languages but we haven't made our final decisions yet for Vista!
Q: [23] Exists the intention to launch codes that can be used for programmers to implement the voice recognition?
A: Yes there will be a set of API's as part of the managed Windows API (WinFX) that will allow you to write speech enabled applications.
Q: [13] Will it be possible to select another language as the localized one? for example if an nativ english speaking person has to use a german vista version for some reason and wants to use english SR instead the default german one.
A: If you want to use the German recognizer, you'll need to set your OS language to German. We understand that many bi-lingual people want to switch back and forth without switching the OS language, but we don't have that feature in Vista. It is a great feature to have, and we hope to have it in a future release.
Q: [35] Will the Speech Recognition technologies in Vista be back ported to Windows XP?
A: Only a developers API for managed code will be avaiable in XP.
Q: [28] speech recognition will be have the power to control enyting of the os like ibm viavoice?
A: You will be able to navigate/control your computer using speech, and you will be able to dictate into documents, emails, etc.
Q: [49] Will there be specific sound cards for speech recognition, or will "onboard sound" be sufficient?
A: For the remaining pre-release versions of Vista, there won't be any dramatic improvements in the TTS engines that's already present, but there is active development going on in TTS that is greatly improving the quality of the voice.
Q: [27] What type of feedback are you looking for from testers for Speech Recognition?
A: We would like to know if speech works for you in Vista both in terms of scenarios and accuracy. Can you use it instead of your keyboard and mouse? More specifically, is the accuracy good enough? what features are missing for you? what do you absolutely love? You can send us feedback directly at listen@microsoft.com
Q: [49] Will there be specific sound cards for speech recognition, or will "onboard sound" be sufficient?
A: There are not specific sound cards for speech. The on-board card should be sufficient
Q: [20] I'm a chinese from Malaysia. I haven't tried the Voice Reconition in Vista, but base on my provious experience in Office 2003 voice reconition, I found that the result is not that accurate, especially in Voica Dictation. I tried English and Chinese version
A: Hi
Q: [15] Speech recognition has been significantly upgraded in Windows Vista; could you give me an overview about what it took since Windows XP to make this major breakthrough?
A: One major difference was our attention to users. In the past I think we thought more about this as a technology, and less about it as a useful tool for people.
For example, we had the ability to spell words in Office XP, but it wasn't a complete experience. You could say "Alpha, Bravo, Charlie", or "A B C" but sometimes you'd get "A. B. C.". And, if one of those letters was wrong, there was no way to go back and correct the letters, other than saying "Press left arrow", "Press left arrow" ...
By starting with users requests, and doing careful usability testing all along the way, we think we've done a much better job delivering what users want in a speech recognition package than we ever have before.
Q: [16] Will speech recognition be able to reproduce individual user natural voices in the near future?
A: That's a great feature. We've thought about that a little, and it's definitely possible.
Q: [48] what applications will be speech enabled? Office 2007? Outlook?
A: In Vista most applications should be accessible via Speech. We take advantage of some of the accessibility framework of the operating system to discover menu items, etc. You should also be able to dictate into most text boxes.
Q: [65] Are MS working on Portuguese language?
A: We are considering addressing speech recognition for the portuguese language. Do not expect to see it in Vista for some time though. There is no time line in place for it.
Q: [18] Surely speech recognition in Vista is a proof of concept idea and is trying to revitalise an underused tech, and to use to properly will will require a third party program?
A: Speech recognition is an application part of Windows Vista. It will not requiere any third party programs to use. You just need a good quality microphone.
Q: [47] will there be a sdk for Speech recognition so programers can add it to there software also
A: Hi Bernie, there is an SDK current, the Speech Application Programming Interface SDK (SAPI) has been available for quite a while now. We also have just added the new managed Speech API, currently known as SpeechFX. It's part of the WinFX ("Avalon" and is in the System.Speech namespace.
Q: [14] Does the Voice works well with non native speakers? Like japanese? Any internal Japanese/Asian testers?
A: Our system is designed to adapt to an individual user's voice. We have colleagues who are native speakers of Asian languages and they are able to use the English speech recognition!
Q: [36] Will German be supported as well?
A: Ja. There is a German recognizer in the February CTP and we are working on the Speech UI for Beta 2.
Q: [22] Will there be a Speech Recognition Gadget?
A: There are no plans currently to build a speech gadget. What would you like to see?
Q: [24] What are the requirements for Speech Recognition? And why would I want to use it?
A: You need a Windows Vista computer and a good quality microphone.
Q: [11] I know this answer could be slightly off topic but I try anyway: I'd like to know if the TTS engine will be further improved in next builds (sound is still very unnatural) and voices in other languages will be added to localized versions.
A: For the remaining pre-release versions of Vista, there won't be any dramatic improvements in the TTS engines that's already present, but there is active development going on in TTS that is greatly improving the quality of the voice.
Q: [25] Could you compare accuracy of Vista ASR and Dragon NS 8.0 from Nuance, please?
A: I think that the accuracy is very similar. In some situations, I believe Windows Speech Recognition is a little more accurate, and sometimes Dragon Naturally Speaking is more accurate. By in large, you should see very similar results.
Q: [46] Is any particular type of headset or mic recommended?
A: Headset USB microphones are preferred. Currently we don't have a list of microphones, but we are working on it.
A: You only need a good quality microphone and Windows Vista. Speech Recognition enables you to dictate email, compose documents, control your favorites applications using your voice and control Windows Vista. It can help you prevent RSI as well.
Q: [55] Will speech recog work better with a dual processor cpu?
A: Hi BJ, yep, speech recognition will work on any hardward configuration supported by Windows Vista.
Q: [26] What languages will be available for Speech?
A: What languages would you like to see?
Q: [81] What is the definition for learning from other documents and e-mail? locations? configurable? is e-mail = Outlook and Microsoft Mail only?
A: We look at the users documents and e-mails as they are provided by the search indexing in Vista. This allows us to adapt to a users vocabulary and the phrase he uses. I am not sure about support of other email programs
Q: [39] whats the status of swedish and finnish support?
A: There are no plans for these two languages at this time.
Q: [30] Post Windows Vista, what do you plan to do to improve Speech Recognition, will there be updates or filters to better improve voice recognition, anything in store?
A: We plan to continue providing updates and tools that wold help developers and end-user to increase accuracy and user experience. Stay tune for downloads from microsoft.com
Q: [30] Post Windows Vista, what do you plan to do to improve Speech Recognition, will there be updates or filters to better improve voice recognition, anything in store?
A: While we will only ship a language if the quality is high, we will continue to actively work to improve the recognition quality of SR for every language even after Vista ships. We are discussing the possibility of delivering updates in service packs.
Q: [33] What are the system requirements to get a good experience using Speech recognition in Windows Vista?
A: Same as minimum requirements for Vista installation.
Q: [31] as a user that has not yet used speech recognition, what do you see as the main uses for speech recognition, will I be able to use the computer without keyboard and/or mouse?
A: Speech recognition can be used to make your interactions with your PC more natural. Some things are best left to the mouse, like drawing in Photoshop, or interacting with some games, like Minesweeper.
Having said that, you should still be able use those programs by voice if you choose, but if you can use a mouse and a keyboard, that might be better in some applications.
However, if you are a slow typist, or just want to lean back in your chair, Speech Recognition could be a good solution.
It's also ideal for people that have mobility impairments such as RSI.
Q: [24] What are the requirements for Speech Recognition? And why would I want to use it?
A: You only need a good quality microphone and Windows Vista. Speech Recognition enables you to dictate email, compose documents, control your favorites applications using your voice and control Windows Vista. It can help you prevent RSI as well.
Q: [63] Is it possible to manually move the mouse, then issue a voice command like "right-click", "double-click" without some extra context to the voice command?
A: You can say "Mousegrid". This will divide the screen into 9 squares. You can move the mouse to the center of a square by saying its number. It will then divide that square in 9 squares again and you can say another number and so on until you get where you need the mouse to be. You can actually say a string of numbers at once if you know already what the sequence is to get you to a particular spot.
Q: [80] Will Bluetooth headsets be supported?
A: Hi Daron, if the headset has a Bluetooth audio profile, it should work with Vista. However, currently BT headsets are limited to 8Khz range and thus the quality of the recognition decreases. This is something that we have improved on Windows Mobiles devices, but not yet with Vista.
Q: [51] I assume you are dogfooding this, how do you use it on a daily basis? What scenarios work the best for you?
A: I've had great success with the activities I do the most: Composing, sending, reviewing email in Outlook, and creating documents in Word.
Q: [91] the speech engine will be more acurate with the external noise?
A: Background noise will allways be an issue. But vista is better at it
Q: [58] Do you have Arabic speach coming?
A: We are not actively working on Arabic for Vista release; however, we are aware of the high demand for this language!
Q: [29] what programs working with speech?
A: Windows Speech Recognition works with most applications written for Windows. There are some exceptions. If an application exposes itself properly to the OS, we will work well with it.
For example, we didn't change Notepad or Wordpad in any ways, but they both work pretty well with speech.
Q: [60] Are you dictating your responses to these questions?
A: I'm not, but that's a great suggestion, I'm sorry I didn't think to setup the chat software on Vista to do that.
Q: [71] Will voice recognition work for UK English as well as US english? This has been an issue with dictation in Word.
A: We are actively working on separate SR for US English and UK English!
Q: [115] I understand that in the past Dragon NaturallySpeaking would NOT WORK with Windows Vista. I know you may wonder why one would need Dragon, when speech recognition is already built into Vista, but I know some will ask.
A: I'm quite confident that DNS will work with Vista once Vista ships. Sometimes ISVs like Nuance have to make small changes in their setup programs, or other areas of their application to work well on a newer OS release.
I know that Nuance is working to make this happen.
Q: [113] will future updates/enhancements need to wait 6 years for the next major Windows release?
A: I hope not.
Q: [119] is there a plan to add a speech to text translator for vista for foreign Language
A: There are no plans for this currently.
Q: [72] there will be a way to add words on speech recognition?
A: There are two ways: In the correction UI you can spell a word that you want to add. You can also record a pronunciation for it then and we will add it to your user lexicon. If you enable document review, then we will look at your documents and emails and add words we don't know to the lexicon as well.
Q: [38] I have not been able to try the speech options on vista yet because of lack of sound card driver support in the CTPs will this get better??
A: The closer we get to RTM we will get better driver support. We are aware of some problems and they been already fixed.
Q: [104] Is there a way to back up your speech profile in Vista and move to another machine?
A: Yes the backup and transfer of speech profiles is taken care of with the file transfer wizard built-in Vista.
Q: [120] Will French, Spanish, German and Chinese Languages be available? Any idea of a range of total languages to be included 10-15?
A: Each of those languages is actively being worked on. Whether they ship or not is still not yet been determined. Many factors need to go into this, one of which is: How accurate is the engine.
Q: [78] Sam: give swag for best questions not random.
A: I take that into consideration, too.
Q: [123] will be a text to speech on spanish?
A: Currently no.
Q: [21] Is there going to be any issue with creative sound blaster 5.1 and speech isues with the drivers?
A: If Vista supports that sound card, it should work with Speech.
Q: [106] In the past MS created nice hardware (I have still my Sidewinder gamevoice in use). Could Microsoft bring up something similar possible on wireless basis to support Speech?
A: Good question Olaf, in the past, Microsoft has created hardware products to promote features of Windows and other products. Another reason is to spur investment by other companies in the same arena. Currently, there are a number of high quality microphones that are available, so I'm not certain that Microsoft would want to enter with it's own product. However, that is just a guess on my part. I will pass your comment to our hardware team.
Q: [37] Will there Developer opportunity with Speech recognition in Windows Vista, to build on top of the current experience to provide more personalized experiences?
A: Yes. You can either user SAPI (a COM based interface), or you can use our new managed code interface: System.Speech.
Q: [50] To use the foregn language versions of speech, will we need to use the MUI's or will they be included either separately or included in Vista?
A: They will be part of the MUI.
Q: [127] I'd like to congratulate the Speech team on the "show numbers", and related abilities to control windows and programs with speech!
A: Thanks! We like it a lot too!
Q: [75] Microphone worked in 5231 but has not in 5270 and 5308. Have followed the suggestions in NGs for setting it up. Are there any issues with soundmax audio that prevent the mic from functioning. Sound is otherwise fine.
A: The entire audio front-end has been rewritten with Windows Vista. I personally use a Toshiba with SoundMAX drivers and they work well in the Febuary CTP (5308). My suggestion would be to ensure that the microphone has the proper role assigned to it and is otherwise selected properly. If you continue to have troubles, please file a bug.
Q: [41] How Speech Recognition taking advantage of some of the new and improved capabilities in Windows Vista, such as MCE, Web browser, will see Speech Recognition using the 10 foot experience that is available with MCE for example?
A: Speech works with any application that exposes it's information via standard accessibility techniques. Both IE and MCE do that.
So ... You should be able to use SR in those apps pretty well.
I know that it works very well in IE:
"Start Internet"
"Go to address"
"w w w dot m s n dot c o m"
"Press enter"
"Scroll down"
"News"
Q: [44] I would like to have option of participating in recognition using the context of my application. Will it be possible?
A: Yes. You can use SAPI or System.Speech.
Q: [127] I'd like to congratulate the Speech team on the "show numbers", and related abilities to control windows and programs with speech!
A: I love this feature! It enables the user to accomplish what they need to on a page even if they're not sure what to say. So it helps you get your tasks done.
Q: [84] will i be able to use speech recognition with another keyboard/regional setting that us/english which is awkward and klunky way just to use the speech recognition
A: You can switch the OS to another local by installing a MUI pack and if there is speech recognition available for it you can then use that language. You cannot use a different language than the current OS language using the Speech UI. You can use a recognizer in a language different from the OS (installed with a MUI pack) inside a program that uses the recognizer in process.
Q: [102] Do you recommend a bluetooth headset for it?
A: No particular recommendations, given the limitations of BT headsets currently (see my other answer on BT headsets). However, I'm a fan of the Jabra 250v for my cell phone and as an occasional headset.
Q: [142] will speech recognition be available at logon?
A: Speech Recognition isn't available on the log on screen itself, but once you've logged on, if you chose to start Speech Recognition at start up, it'll be there ready for you to use.
Q: [146] Is Japanese speech recognition supported in the just released Japanese CTP?
A: Japanese speech recognition will be supported in Beta 2
Q: [135] any way to eliminate some of the background noise, or improve the system to recognize your voice over the rest of noise?
A: You can adapt to your change of environments by running the microphone wizard again. You can improve accuracy by taking additional voice trainings. Noise canceling microphones also help remove background noise.
Q: [59] About Q: [8], are you working to support brazilian portuguese?
A: We are not actively working on Brazilian Portuguese for Vista, but we are aware of significant interest for this language.
Q: [82] Can you bring voice training files from XP into Vista? What about training files from Dragon?
A: XP to Vista
Q: [34] Anyone with MS gone to on-line banking and used speec rec to input passwords?
A: Using speech recognition for passwords is more difficult than for other text fields. You'll likely have to use keyboard emulation. Are you having a specific problem, or a issue with a particular site?
Q: [67] will the speach API's be available in Longhorn server in the future, if so could they be used for telephony applications al la MS Speech server ?
A: SAPI and System.Speech will be there. The recognizers won't be installed, though. You'd have to obtain a recognizer separately.
Q: [154] How many custom words can be added to the lexicon, is it an unlimited amount?
A: At this time 1000 words can be added. Do you need more?
Q: [52] Hi, will it be possible to have various languages to select in the final release? I'd like to have all languages available, or at least as free downloads.
A: If you can run your installation of Windows Vista in a particular language that we support, SR will be avaible.
Q: [112] Will there be a differenation between English-US and English-UK for speech?
A: We consider English-US different enough from Enflish-UK to warrant a different recognizer for it. We are actively working on UK-English.
Q: [56] And will the performance be much more increased (or the needed resources be lowered) in the final release?
A: The resources needed in the final release will be very similar to that needed now.
Q: [169] I SR is available for Media Center can this also be used on an XBox 360 and use the 360 headset for control???
A: Not currently.
Q: [111] You mentioned USB headset/mic, does that work better than a mic with soundcard, or did I misunderstand your ans
A: I think *in general* USB headsets or microphones work better, but that's a subjective opinion. It really depends on the quality of the components used. Generally the USB headsets are highly quality than the cheapy headsets that have separate mic and earphone plugs. But if you have a good quality headset and good quality audio hardware, you should get good performance with speech recognition.
Q: [159] which speech app is MS using for Speech Recognition (Learnhaut-Hagtberg??) or dragon you-know!
A: We are using our own speech recognizer that has come out of years of R&D at Microsoft and that is a continuation of the recognition engine that was in Office 2003 for example. We redesigned our own Speech User Interface from the ground up for Vista.
Q: [167] Speech has always been pretty good! How is MS trying to get MORE ppl to utilize it?
A: We're making it better.
More people will use it once it's significantly easier to use, and available in more languages.
From a marketing perspective, we haven't finalized our plans yet.
Q: [121] Are there any plans to integrate Text-To-Speech natively in common Windows controls such as text boxes? Such feature already exists in Apple Mac OS.
A: For speech output, Microsoft has included Narrator in all versions of Windows for several years now. It will speak the name of controls as they are selected and/or focused.
Q: [161] Does Speech Recognition support Media Center?
A: In some ways, yes. Although, not in the coolest ways like "Record Survivor at 8 PM on Thursday". Although, you could add a macro to do such a recording if you'd like with a future Macro tool that we're creating.
Q: [133] Will you need to address the voice recognition feature somehow? Or in a different perspective of this, will talking nearby affect or activate the voice recognition feature?
A: User is always in control. Once you have Speech Running you can say 'stop listening' and the Speech recognition application goes into sleep mode. It can be awaken by saying 'Start listening'. If you have a mute button on your microphone, you can also use that to control when the speech application is listening.
Q: [162] Are you going to develop some more advanced API oriented to the processing text from dictation mode (syntax parsing, semantics)? So we will be able to build truly voice interface. And not to make user remember the list of predefined commands
A: I'd like to hear more about what you'd like to see developed. You should send us some mail to follow up offline at listen@microsoft.com.
Q: [155] Who is the main target for speech recog. in vista. In my business, the medical and legal professions have the greatest for this. Is there any special attention given to certain users based on the flavor of the OS ?
A: Vista is shipping with features that could be enable in the future once we get a chance to provide tools to developers to support specialized languages such as medical and law. This is still in planning stage. But the OS will support it. Microsoft does not plan to provide the vocabularies needed. It would likely require ISV's to do so.
Q: [185] Many severely disabled users control the computer entirely by voice. They do not have the option of using the keyboard to type in login credentials. If speech is not available to login, have any alternatives been considered for disabled?
A: What do they use today?
Q: [171] Will Speech in Vista support similar "read back" features in Outlook for PIM info, similar to MS Voice Command on the WinMobile platform? If not, could this be added as a feature request?
A: It's not currently supported, but I'd really like that feature too. Again, a Macro could be created to do this.
Q: [156] Wil SR be available on Longhorn Server, and if so via a Terminal Server session?
A: Speech Recognition doesn't work via Terminal Server currently.
Q: [99] On a somewhat related note, we have customers using the Philips SpeechMike Pro (serial interface) and who want to switch to the USB version. Will they have any problems with compatiblity?
A: They *should* not have problems with the USB versions, but it's highly dependent on the drivers that Philips provides. Today I'm expecting a shipment of microphones from them for testing purposes so I'll find out for myself. Do you know the model number of the mic they plan on using?
Q: [172] Will the user be able to wake up his PC from Standby using Speech?
A: Not at this time. That would be a great feature though.
Q: [82] Can you bring voice training files from XP into Vista? What about training files from Dragon?
A: For XP to Vista upgrade, only user added words will be used for the new profile. Users will have to re-do training. Dragon training files will be transfered on XP to Vista upgrade, but they will not be used by our recognizer. If you are doing clean install than Dragon should have some kind of profile and data transfer tool/system.
Q: [163] My understanding is bluetooth has a lower quality output than other wireless methods, so would not be suitable for voice recognition, is that correct?
A: The Bluetooth standard currently only allows recording in 8kHz (vs. 16kHz typically used for speech recognition on the desktop). This was mainly geared towards mobile phone use scenarios. So we would expect some degradation in accuracy when using a bluetooth microphone due to the lower quality signal.
Q: [186] Do you expect Windows Vista speech recognition accuracy to significantly improve before Vista ships or is what we have now essentially what we should expect in the shipping version of the OS?
A: It will continue to increase, but probably not substantially.
Q: [126] There are several version of Dragon naturally Speaking - one of the main that I have used is the pro- legal and medical what are Vista's speech recognition's capabilities in that region
A: You certainly can write code with speech recognition, but given the amount of symbols, it's tedious at best. I recently read that there are a number of software developers here at Microsoft who use speech recognition on a full or part time basis.
Q: [76] Do you think speech recognition might correct enunciation of some of our teens?
A: Maybe! Correct enunciation will definitively boost the accuracy of dictation.
Q: [69] I wanted to try out the speech functionality, but I've been unable to run Vista Beta 5308. When installing under Virtual PC 2004, I get the error "Windows Setup: setup.exe - Application Error The exception unknown software exception (0xe0000100)".
A: Please log that but if you h aven't already.
Q: [194] Will speech recognition work over remote desktop?
A: Not currently.
Q: [189] main english US version word engines Webster?
A: We use our own lexicon that's been developed for the last 10+ years.
Q: [181] can this be used with external devices to control media center devices, home automation systems, security systems?
A: You could certianly write such a program using SAPI or System.Speech.
Q: [130] Follow up to 71. Are we only seeing US English currenlty in Vista builds?
A: As far as OS languages are concerned, we are releasing German and Japanese as well as English. We will be releasing more languages around beta 2, but aren't ready to announce exactly which ones yet.
Q: [196] Many severely disabled users must have assistants or co workers log in on their behalf. It would interesting to know the position of Microsoft's Accessiblity Division on this issue. What has been their role in the development of speech recognition?
A: We work closely with the Accessibility Technology Group (ATG). Although we're not supporting running Speech Recognition on the log on screen in Vista, it is a goal someday.
Q: [92] Will there be an option for users to export their voice training profiles so they can take it with them to different PCs
A: Yes. You will be able to use "Windows Easy Transfer" under System tools to move speech profiles (voice training data) and other documents and setting from one PC to the other.
Q: [145] Do you know of any 3rd party vendors taking advantage of the newer API to make custom 'answering machine' / 'voicemail' systems with the new functionalities? You could go pretty gonzo if you really wanted to for that type of thing.
A: For telephony scenarios like this one we offer Microsoft Speech Server as a platform and there is a whole partner ecosystem working on various applications in the telephony space. We are also integrating this functionality into Exchange 12 and this will allow you to interact with your emails and voicemails via speech.
Q: [148] Has the introduction of the UMPC plaform influenced changes in the Speech abilities of Vista? It seems like this new platform could really benifit from speech based control.
A: It hasn't influenced it directly, no, but we've been working with the UMPC team for quite some time.
Q: [184] I hope this question is not redundant in this fast and furious chat: Many doctors hope to to use the native VR in Vista. Will that be addressed with more specificity and complexity, with respect to medical vocabularies etc.?
A: The native VR in vista will not do a good job with specialized llanguages such as Medical, Law, Engineering, etc. Medical fields such as radiology and others required specialized vocabularies and/or Dictation Topics (also known as Language models). Vista will be able to use specialized language models other than the one provided natively. What remains is for MS to create the tools needed to enable this feature. This is something we are still at a planning stage. But we are certatinly interested. I would like to hear more from you on this topic.
Q: [110] Will speech recognition detect voices of different people and respond according kind of like individual "voice profiles"? Will I be able to have "voice passwords" that only my voice can execute?
A: Currently the speech profile is keyed to the logged on user. Let's say that Bob is the currently logged on user and he has setup and trained the speech profile and uses it regularly. If his daughter Barbara starts using speech recognition, under the Bob user account, the quality of the recognition will be poor. Also, the speech profile will, over time, attempt to adapt to the new voice and thus make the recognition for Bob poor as well when he uses it. So, it's not recommended to have multiple people using the same profile.
Q: [90] Speech recognition seems to require certain words to be said with an american dialect or say them in an american way, any comments on this?
A: We train our acoustic models for American English with a large speech corpus designed to represent the range of pronunciations found in American English. We provide two features to make SR better for each individual user: 1) Speaker adaptation. The speaker-independent engine shipped with the OS rapidly adapts to an individual's accent, and becomes a speaker-dependent engine. 2) Adding pronunciations. Users can add new words and new pronunciations to the system.
Q: [94] Will their be a Russian speech recognition? I go there a couple times a year on Missions riips and it would be good.
A: Sorry, currently we don't have plans on shipping Russian speech recognition.
Q: [179] Q154: We need to be able to add at least 200,000 words from a medical dictionary to the lexicon.
A: For medical situations, you really need a full blown lanuage model to predict the relative frequency of words occurring in sequences. We are working to make that be available, but that's separate from adding user defined words to the lexicon.
Q: [177] Follow up to 154, is this per user or system total? Per user this figure sounds ok, but total system it may be low.
A: Per user. You can actually add more, but the recognizer stops considering them after the 1st 1000. I'll take this feedback back to the development team and see if there's something we can do to make it better.
Q: [212] How about transferring ViaVoice user data to Vista? Will this be possible?
A: That's no supported.
Q: [107] If I have only a USB mike, how can I rig it to work properly (vs. headset)?
A: Your USB mic should be fine to work with speech recognition. Make sure Vista recognizes it properly and that it is set a the default input device to be used with speech recognition.
Q: [124] Will we have to manually switch between languages or will it auto detect the language being spoken?
A: For the Speech UI in Vista you will only be able to use the languge of the operating system. Applications can run the recognizer in process and can switch between different recognition engines. The application could be designed to switch to another language based on a command for example. There is no explicit language detection functionality as such.
Q: [199] how SR perform www.anypage.com, sr support dictation of uRL's
A: In vista we are improving the way we recognize URL's. Dictating then should work in Vista.
Q: [54] Is this something that might be usefull for the blind? Will there be books read by the home computer and computer commands available with voice commands?
A: Speech recognition is useful for people who need alternative means of input. Some people who are blind also have those needs. Rather, the other tools that Microsoft provides are more suited for visual impairments, such as Narrator and Magnifier. To be aware those, that these tools are *very basic* and are meant to bootstrap users into the platform. There are many companies devoted to providing a much better experience for people who have low vision or are totally blind. See http://microsoft.com/enable/ for more information.
Q: [214] will vista speech be 508 compatible?
A: Yes, it should be.
Q: [204] Will Voice Recognition work the next build? Launching it currently crashes rundll on all of my systems, and yes it's been bugged
A: It should work much better in Beta 2.
Q: [198] Does Speech recognition have a built in translator?
A: No not at this time.
Q: [70] will the speec engine integrate and transcribe from digital voice recorders
A: We won't support that feature out of the box. An application could be written to that though, using SAPI or System.Speech.
Q: [95] What are the languages (besides English, German and Japanese) that will be supported by Speech Recognition; and how can users 'install' definition files for other languages?
A: To install other languages supported by Speech recognition, you will need to install MUI packs and switch OS to that language locale. Than you will be able to start Speech Recogniton for that language.
Q: [79] I am very curious to how speech works with office 12 is it the same with same function as it is in Office 2003
A: Office 12 actually removed all of their native speech support, but when you run Office 12 on Vista, you should get an even better experience than what you had in Office 2003.
Q: [223] Can SR be used to switch between apps?
A: Yes you can use the command "Switch to <application_name>" to switch between running apps.
Q: [68] will speach recognition be integrated with intelasense?
A: Good question RLM, there is nothing specific in the Speech UX for IntelliSense currently. Can you give an example of how you'd like it to work?
Q: [85] Is there a guide to use for trying speech recognition, perhaps a URL, to get us staarted?
A: You could take a look at this screen-cast. It has a good introduction to what you can do with Speech Recognition in Vista. http://blogs.technet.com/chenley/archive/2006/02/21/420136.aspx
Q: [202] Will VOIP be a a possible input source, a la WM5.0 devices?
A: No for the main user experience but you can write an application to do it.
Q: [101] re [46], why would an USB mic be better than a analogue one?
A: Yes. You can also use your analog microphone with USB Pod.
Q: [114] I used dragon naturally speaking professional there is a training session in each program its suppose to be totally hands free , can you train speech recognition to add on new words its not familiar with or is it just a light program
A: We provide a totally hand free tutorial/training session as well as additional voice trainings with Windows Vista. You can also use the speech dictionary feature to add and train new words. We also have a document review feature that scans your documents/emails if you want and picks up the words you frequently use. Are you expecting more features in this area?
Q: [229] is there a total number of words you can add to the vocabulary? Can you add a list (an ASCII listing), of words directly to the vocabulary.
A: 1000 words max, currently. We don't have UI to add words in bulk at the moment, but it's a good feature suggestion. An applictation could easily be written by a 3rd party to do this.
Q: [221] do you think voice recognition could help you to get that data and translate it to other languages??
A: Translation is not a feature in Vista. It is conceivable though that a 3rd party could use the speech recognition capability in Vista for one of the supported languages as an input to their translation software.
Q: [85] Is there a guide to use for trying speech recognition, perhaps a URL, to get us staarted?
A: Try this:
http://www.microsoft.com/technet/windowsvista/library/c208e792-e591-455a-82d9-a98264324e0d.mspx
and this:
http://blogs.technet.com/chenley/archive/2006/02/21/420136.aspx
Q: [213] Will very specific professional terms of VR dictation supported in the future? For example spino-cerebeller degeneration in medical?
A: Vista native dictation support is for standard English more like the one you find in news papers for the general public and business community. For medical, the system requires a specialize dictation topic to be created and installed. Vista will support such thing but MS would need to provide the tools required to build them. We are still at a planinig stage for this. We would appreciate your feedback
Q: [128] This is not exactly about speech recognition, but also a speech-like feature: Why does the "Narrator" have a female voice now? I liked the male voice (which was called Sam in 2000/XP) better.
A: The reason is that we changed the TTS engines in Vista to reflect new technology. Microsoft Anna is a step above what was previouly available. I cannot say with certainty that a male counterpart will ship in the final release of Vista.
Q: [236] re [101] (and [46]) are there any technical reasons for why a usb mic (digital) would be better than an analogue one? If that's the case does that not indicate that your A/D conversion suck?
A: The A/D conversion is done in hardware. Thus, because AC97 hardware is on the motherboad, there is sometimes electrical noise introduced. With USB, the A/D is done outboard and thus less prone to that noise.
Q: [83] Is there a point to use speech in a single room office with many people around?
A: Three things come to mind here: 1) The main challenge of that scenario is capturing sufficient quality audio from the multiple speakers - quality microphones and setup to provide a reasonable signal to the SR. 2) The speech recognizer adapts to each user to improve recognition accuracy, and links the adaptation information to that user's profile. A profile for one user may or may not work as well for another user. 3) Multiple speakers on a single audio feed produce overlapping waveforms. SR is designed for a single speech stream. I can envision a future when multiple parties in a meeting could each have their own mobile device capturing their speech (speaker-adapted recognition, good quality audio input), which is then synched up with the rest of the meeting!
Q: [187] Will the system be able to read back ASCII books, interpreting punctuation correctly?
A: Narrator and our TTS engines do a excellent job of speaking even complex phrases
Q: [116] What apps will natively work with Vista speech recognition; Office, OneNote, IE, others?
A: Most applications will work with speech recognition!
Q: [131] Will speech also work in a way, that you maybe start a program by speech command the program is reporting back via speech output "Program xyz is ready now"?
A: You could write a macro that does this.
Q: [136] Will there be a Vista Logo Programme for headsets for SR?
A: Not specifically for speech recognition, but the Designed for Windows logo program is available for hardware. You can be sure that if a device has the Designed for Windows Vista logo, it will work with Speech.
Q: [224] Have you guys made any progress towards having multiple SR applications be able to share the same feedback and commands windows? Or does each application that uses SR have to provide its own?
A: Hi Mike. The feedback window is programmable. You can send your text feedback there. The What Can I Say window is still not configurable by ISVs.
Q: [237] I've noticed that sometimes when I play music on my computer loud enough, SR picks up the noise. Is SR going to be able to block out the noise coming from that same computer?
A: Doing echo cancellation to remove music playing from the computer in the background is not a feature at this point. To get the best recognition accuracy I would recommend that you either mute or at least turn down the music.
Q: [225] Scenario: I have a MUI system and use German as my default language, train some German words. Then I switch to a different language, say Spanish - will my training and my custom words be remembered when I switch back to German?
A: Yes they will
Q: [230] como configuro o windows vista para usar o memory key para extender a memoria ram?
A: Hola esta pregunta no es acerca de Speech Reconigtion. I can't answer
Q: [218] SR in vista is great, it's one of the best SR's I've used, what are the plans to "advertise" sr? As in make the average user know it's available.
A: Thanks. Please tell us what you like at listen@microsoft.com The Windows Vista marketing team is working on that. We have a web presence already here: http://www.microsoft.com/windowsvista/features/foreveryone/speech.mspx
Q: [105] What's the recommended course of action when implementing a command in an application that conflicts with the systems built in commands?
A: Vista's speech APIs have a prioritization mechanism. The application can choose to to override commands, or some commands might be overriden by the SR itself.
Beyond that, it depends on the command. If you want to follow up, feel free to send email to listen@microsoft.com and we can discuss it offline.
A: We have similar accuracy.
Q: [97] Suggestion: What about using speech as a password/log on in the future? It will detect the voices of us as an alternative to typing our password
A: Jabez, that is an excellent suggestion and it's something that our security people would love to see happen as part of a multi-level authenication scheme. However, given the challenges of just getting it right on the desktop, it'll be awhile before we can "speech enable" the login process.
Q: [143] you say you "will look at your documents and emails and add words we don't know to the lexicon". Will the user have some control over that? Which documents to harvest? Which harvested words to accept?
A: We use the same system that the indexer in Vista uses. So, if you can use Search to find the document, Speech will consider it when it updates the language models.
Q: [139] Will there be any special add-ons for the speech dictionary to include things like special medical and legal terms?
A: Vista native dictation support is for standard English more like the one you find in newspapers for the general public and business community. For medical, the system requires a specialize dictation topic to be created and installed. Vista will support such thing but MS would need to provide the tools required to build them. We are still at a planinig stage for this. We would appreciate your feedback
Q: [176] will speech be smart enough to "stop listening" if i get a call on a softphone or voice chat?
A: Not currently, but that's a great suggestion.
Q: [9] Is a noise cancelling mic necessary for good results?
A: A good quality mic will help with recognition accuracy. Noise cancellation might help in somewhat noisy environments, but probably won't make a noticable difference if you use it in a quite environment.
Q: [178] When developing applications that make use of speech recognition in Vista, will the software logo requirement require System.Speech, or may SAPI be used as well?
A: You can use either one.
Q: [158] In an Enterprise environment, will we be able to control how much control or who gets access to speech?
A: Yes. You could lock down the registry location we configure SR to automatically start.
Q: [174] Say please, probability is which that there will be support of Russian in Beta2.
A: We are not actively working on Russian for Beta 2.
Q: [205] I have used the speech typing tool in XP, it took a really long time to get it set up, Will Vista be the same? Is there a way to import the file from XP to Vista?
A: No, the training process is more straight forward, and is intergrated with a tutorial on how to use the system. No, you cannot import the speech profile from a previous version - even if you could, it wouldn't help much - the recognition in Vista is much better even without a trained profile.
Q: [245] Do I understand correctly that you can NOT currently use voice to "open word" "save" "print" "close word" and shut down the computer, stuff like that? Is that something that is coming?
A: We DO support in Vista the commands you are mentioning above. It is already in the product.
Q: [242] could you know the msagent team ?
A: We also own MSAgent.
Q: [149] How well will speech integrate with other Microsoft and 3rd Party applications without to much user hassles?
A: There are two level of API's one for COM based applications (SAPI) and one for Managed code. If you have a specif quetsion on what exactly you want to integrate please ask. Thanks.
A: I am not sure about your question, but USB microphones are OK. The difference will be USB vs. Analog, and we recomend USB.
Q: [246] Will there be a "push button to listen" option where voice will only be listening when I hold a button down?
A: Sort of. "Ctrl-Windows" turns speech on and off. Would you like to see push to talk added? What about configuring the key?
Q: [231] Are there any serach specific special commands exposed in Vista SR?
A: No but you can control the search UI via voice.
Q: [144] Will lexicon expansions be made available to the OS via Windows Update?
A: Yes, SAPI allows for two levels of lexicons, Application and User level lexicon
Q: [245] Do I understand correctly that you can NOT currently use voice to "open word" "save" "print" "close word" and shut down the computer, stuff like that? Is that something that is coming?
A: You CAN currently do all those things with Speech. You should be able to say any command (menu item, dialog box, etc.) you can see on the screen. Rob Chambers has a bunch of commands listed in his blog.
Q: [228] Narrator crashes occasionaly, whould that affect speech input?
A: It should not affect speech input. Please, please, please file bugs against any crashes
Q: [244] Is microsoft working with computer/monitor manufacters so that microphones will be built into the hardware?
A: Actually, built in mics in monitors and laptops usually do poorly with speech recognition. A high quality microphone is your best bet.
Q: [100] Could you compare accuracy of Vista ASR and viavoice, please?
A: Accuracy is similar
Q: [87] [14] Lesley: Sure your colleagues can use English SPeech recognition, since they all are living in the States! What about your colleagues working from Japan/China with no american accent?
A: True, but they actually demonstrate a wide range of American English influence in their phonologies. You are right to point out that ideally, models would exist for varieties of English beyond US and UK, such as Singapore English, etc.
Q: [9] Is a noise cancelling mic necessary for good results?
A: A good microphone will definitively improve the quality of recognition, especially in dictation. Noise cancelling microphones will definitively boost your accuracy in a moderate noisy environments.
Q: [6] will there be any conflicts with running Dragon Natural, or is there a gap between the two that would even require the need for it
A: There shouldn't be any conflicts between Dragon Naturally Speaking and Windows Vista Speech Recognition. You should be able to run either one on Windows Vista.
Q: [2] speech recognition will be on spanish language ?
A: Spanish is one of the languages we are actively working on for Vista. We are simultaneously working on several languages, but we haven't made our final decisions on which languages will be shipped with Vista.
Q: [3] Does the french recognition coming ?
A: French is also a language we are actively working on for Vista
Q: [12] I've been afraid to use voice for a number of reasons. 1. it doesn't really work very well and 2. it really uses a lot of computer resources and can cause a lot of problems. Is this still true?
A: We've tried really hard in Windows Vista to make it easier to use and to be much more accurate. We hope that you've seen some of that in the CTPs and in Beta 1.
Q: [7] Is Polish recognition coming to Windows some day?
A: We don't have Polish scheduled in the short term. Thank you for that suggestion!
Q: [8] For how many languages the Microsoft intends to launch the system of voice recognition?
A: We are actively working on several languages but we haven't made our final decisions yet for Vista!
Q: [23] Exists the intention to launch codes that can be used for programmers to implement the voice recognition?
A: Yes there will be a set of API's as part of the managed Windows API (WinFX) that will allow you to write speech enabled applications.
Q: [13] Will it be possible to select another language as the localized one? for example if an nativ english speaking person has to use a german vista version for some reason and wants to use english SR instead the default german one.
A: If you want to use the German recognizer, you'll need to set your OS language to German. We understand that many bi-lingual people want to switch back and forth without switching the OS language, but we don't have that feature in Vista. It is a great feature to have, and we hope to have it in a future release.
Q: [35] Will the Speech Recognition technologies in Vista be back ported to Windows XP?
A: Only a developers API for managed code will be avaiable in XP.
Q: [28] speech recognition will be have the power to control enyting of the os like ibm viavoice?
A: You will be able to navigate/control your computer using speech, and you will be able to dictate into documents, emails, etc.
Q: [49] Will there be specific sound cards for speech recognition, or will "onboard sound" be sufficient?
A: For the remaining pre-release versions of Vista, there won't be any dramatic improvements in the TTS engines that's already present, but there is active development going on in TTS that is greatly improving the quality of the voice.
Q: [27] What type of feedback are you looking for from testers for Speech Recognition?
A: We would like to know if speech works for you in Vista both in terms of scenarios and accuracy. Can you use it instead of your keyboard and mouse? More specifically, is the accuracy good enough? what features are missing for you? what do you absolutely love? You can send us feedback directly at listen@microsoft.com
Q: [49] Will there be specific sound cards for speech recognition, or will "onboard sound" be sufficient?
A: There are not specific sound cards for speech. The on-board card should be sufficient
Q: [20] I'm a chinese from Malaysia. I haven't tried the Voice Reconition in Vista, but base on my provious experience in Office 2003 voice reconition, I found that the result is not that accurate, especially in Voica Dictation. I tried English and Chinese version
A: Hi
Q: [15] Speech recognition has been significantly upgraded in Windows Vista; could you give me an overview about what it took since Windows XP to make this major breakthrough?
A: One major difference was our attention to users. In the past I think we thought more about this as a technology, and less about it as a useful tool for people.
For example, we had the ability to spell words in Office XP, but it wasn't a complete experience. You could say "Alpha, Bravo, Charlie", or "A B C" but sometimes you'd get "A. B. C.". And, if one of those letters was wrong, there was no way to go back and correct the letters, other than saying "Press left arrow", "Press left arrow" ...
By starting with users requests, and doing careful usability testing all along the way, we think we've done a much better job delivering what users want in a speech recognition package than we ever have before.
Q: [16] Will speech recognition be able to reproduce individual user natural voices in the near future?
A: That's a great feature. We've thought about that a little, and it's definitely possible.
Q: [48] what applications will be speech enabled? Office 2007? Outlook?
A: In Vista most applications should be accessible via Speech. We take advantage of some of the accessibility framework of the operating system to discover menu items, etc. You should also be able to dictate into most text boxes.
Q: [65] Are MS working on Portuguese language?
A: We are considering addressing speech recognition for the portuguese language. Do not expect to see it in Vista for some time though. There is no time line in place for it.
Q: [18] Surely speech recognition in Vista is a proof of concept idea and is trying to revitalise an underused tech, and to use to properly will will require a third party program?
A: Speech recognition is an application part of Windows Vista. It will not requiere any third party programs to use. You just need a good quality microphone.
Q: [47] will there be a sdk for Speech recognition so programers can add it to there software also
A: Hi Bernie, there is an SDK current, the Speech Application Programming Interface SDK (SAPI) has been available for quite a while now. We also have just added the new managed Speech API, currently known as SpeechFX. It's part of the WinFX ("Avalon" and is in the System.Speech namespace.
Q: [14] Does the Voice works well with non native speakers? Like japanese? Any internal Japanese/Asian testers?
A: Our system is designed to adapt to an individual user's voice. We have colleagues who are native speakers of Asian languages and they are able to use the English speech recognition!
Q: [36] Will German be supported as well?
A: Ja. There is a German recognizer in the February CTP and we are working on the Speech UI for Beta 2.
Q: [22] Will there be a Speech Recognition Gadget?
A: There are no plans currently to build a speech gadget. What would you like to see?
Q: [24] What are the requirements for Speech Recognition? And why would I want to use it?
A: You need a Windows Vista computer and a good quality microphone.
Q: [11] I know this answer could be slightly off topic but I try anyway: I'd like to know if the TTS engine will be further improved in next builds (sound is still very unnatural) and voices in other languages will be added to localized versions.
A: For the remaining pre-release versions of Vista, there won't be any dramatic improvements in the TTS engines that's already present, but there is active development going on in TTS that is greatly improving the quality of the voice.
Q: [25] Could you compare accuracy of Vista ASR and Dragon NS 8.0 from Nuance, please?
A: I think that the accuracy is very similar. In some situations, I believe Windows Speech Recognition is a little more accurate, and sometimes Dragon Naturally Speaking is more accurate. By in large, you should see very similar results.
Q: [46] Is any particular type of headset or mic recommended?
A: Headset USB microphones are preferred. Currently we don't have a list of microphones, but we are working on it.
A: You only need a good quality microphone and Windows Vista. Speech Recognition enables you to dictate email, compose documents, control your favorites applications using your voice and control Windows Vista. It can help you prevent RSI as well.
Q: [55] Will speech recog work better with a dual processor cpu?
A: Hi BJ, yep, speech recognition will work on any hardward configuration supported by Windows Vista.
Q: [26] What languages will be available for Speech?
A: What languages would you like to see?
Q: [81] What is the definition for learning from other documents and e-mail? locations? configurable? is e-mail = Outlook and Microsoft Mail only?
A: We look at the users documents and e-mails as they are provided by the search indexing in Vista. This allows us to adapt to a users vocabulary and the phrase he uses. I am not sure about support of other email programs
Q: [39] whats the status of swedish and finnish support?
A: There are no plans for these two languages at this time.
Q: [30] Post Windows Vista, what do you plan to do to improve Speech Recognition, will there be updates or filters to better improve voice recognition, anything in store?
A: We plan to continue providing updates and tools that wold help developers and end-user to increase accuracy and user experience. Stay tune for downloads from microsoft.com
Q: [30] Post Windows Vista, what do you plan to do to improve Speech Recognition, will there be updates or filters to better improve voice recognition, anything in store?
A: While we will only ship a language if the quality is high, we will continue to actively work to improve the recognition quality of SR for every language even after Vista ships. We are discussing the possibility of delivering updates in service packs.
Q: [33] What are the system requirements to get a good experience using Speech recognition in Windows Vista?
A: Same as minimum requirements for Vista installation.
Q: [31] as a user that has not yet used speech recognition, what do you see as the main uses for speech recognition, will I be able to use the computer without keyboard and/or mouse?
A: Speech recognition can be used to make your interactions with your PC more natural. Some things are best left to the mouse, like drawing in Photoshop, or interacting with some games, like Minesweeper.
Having said that, you should still be able use those programs by voice if you choose, but if you can use a mouse and a keyboard, that might be better in some applications.
However, if you are a slow typist, or just want to lean back in your chair, Speech Recognition could be a good solution.
It's also ideal for people that have mobility impairments such as RSI.
Q: [24] What are the requirements for Speech Recognition? And why would I want to use it?
A: You only need a good quality microphone and Windows Vista. Speech Recognition enables you to dictate email, compose documents, control your favorites applications using your voice and control Windows Vista. It can help you prevent RSI as well.
Q: [63] Is it possible to manually move the mouse, then issue a voice command like "right-click", "double-click" without some extra context to the voice command?
A: You can say "Mousegrid". This will divide the screen into 9 squares. You can move the mouse to the center of a square by saying its number. It will then divide that square in 9 squares again and you can say another number and so on until you get where you need the mouse to be. You can actually say a string of numbers at once if you know already what the sequence is to get you to a particular spot.
Q: [80] Will Bluetooth headsets be supported?
A: Hi Daron, if the headset has a Bluetooth audio profile, it should work with Vista. However, currently BT headsets are limited to 8Khz range and thus the quality of the recognition decreases. This is something that we have improved on Windows Mobiles devices, but not yet with Vista.
Q: [51] I assume you are dogfooding this, how do you use it on a daily basis? What scenarios work the best for you?
A: I've had great success with the activities I do the most: Composing, sending, reviewing email in Outlook, and creating documents in Word.
Q: [91] the speech engine will be more acurate with the external noise?
A: Background noise will allways be an issue. But vista is better at it
Q: [58] Do you have Arabic speach coming?
A: We are not actively working on Arabic for Vista release; however, we are aware of the high demand for this language!
Q: [29] what programs working with speech?
A: Windows Speech Recognition works with most applications written for Windows. There are some exceptions. If an application exposes itself properly to the OS, we will work well with it.
For example, we didn't change Notepad or Wordpad in any ways, but they both work pretty well with speech.
Q: [60] Are you dictating your responses to these questions?
A: I'm not, but that's a great suggestion, I'm sorry I didn't think to setup the chat software on Vista to do that.
Q: [71] Will voice recognition work for UK English as well as US english? This has been an issue with dictation in Word.
A: We are actively working on separate SR for US English and UK English!
Q: [115] I understand that in the past Dragon NaturallySpeaking would NOT WORK with Windows Vista. I know you may wonder why one would need Dragon, when speech recognition is already built into Vista, but I know some will ask.
A: I'm quite confident that DNS will work with Vista once Vista ships. Sometimes ISVs like Nuance have to make small changes in their setup programs, or other areas of their application to work well on a newer OS release.
I know that Nuance is working to make this happen.
Q: [113] will future updates/enhancements need to wait 6 years for the next major Windows release?
A: I hope not.
Q: [119] is there a plan to add a speech to text translator for vista for foreign Language
A: There are no plans for this currently.
Q: [72] there will be a way to add words on speech recognition?
A: There are two ways: In the correction UI you can spell a word that you want to add. You can also record a pronunciation for it then and we will add it to your user lexicon. If you enable document review, then we will look at your documents and emails and add words we don't know to the lexicon as well.
Q: [38] I have not been able to try the speech options on vista yet because of lack of sound card driver support in the CTPs will this get better??
A: The closer we get to RTM we will get better driver support. We are aware of some problems and they been already fixed.
Q: [104] Is there a way to back up your speech profile in Vista and move to another machine?
A: Yes the backup and transfer of speech profiles is taken care of with the file transfer wizard built-in Vista.
Q: [120] Will French, Spanish, German and Chinese Languages be available? Any idea of a range of total languages to be included 10-15?
A: Each of those languages is actively being worked on. Whether they ship or not is still not yet been determined. Many factors need to go into this, one of which is: How accurate is the engine.
Q: [78] Sam: give swag for best questions not random.
A: I take that into consideration, too.
Q: [123] will be a text to speech on spanish?
A: Currently no.
Q: [21] Is there going to be any issue with creative sound blaster 5.1 and speech isues with the drivers?
A: If Vista supports that sound card, it should work with Speech.
Q: [106] In the past MS created nice hardware (I have still my Sidewinder gamevoice in use). Could Microsoft bring up something similar possible on wireless basis to support Speech?
A: Good question Olaf, in the past, Microsoft has created hardware products to promote features of Windows and other products. Another reason is to spur investment by other companies in the same arena. Currently, there are a number of high quality microphones that are available, so I'm not certain that Microsoft would want to enter with it's own product. However, that is just a guess on my part. I will pass your comment to our hardware team.
Q: [37] Will there Developer opportunity with Speech recognition in Windows Vista, to build on top of the current experience to provide more personalized experiences?
A: Yes. You can either user SAPI (a COM based interface), or you can use our new managed code interface: System.Speech.
Q: [50] To use the foregn language versions of speech, will we need to use the MUI's or will they be included either separately or included in Vista?
A: They will be part of the MUI.
Q: [127] I'd like to congratulate the Speech team on the "show numbers", and related abilities to control windows and programs with speech!
A: Thanks! We like it a lot too!
Q: [75] Microphone worked in 5231 but has not in 5270 and 5308. Have followed the suggestions in NGs for setting it up. Are there any issues with soundmax audio that prevent the mic from functioning. Sound is otherwise fine.
A: The entire audio front-end has been rewritten with Windows Vista. I personally use a Toshiba with SoundMAX drivers and they work well in the Febuary CTP (5308). My suggestion would be to ensure that the microphone has the proper role assigned to it and is otherwise selected properly. If you continue to have troubles, please file a bug.
Q: [41] How Speech Recognition taking advantage of some of the new and improved capabilities in Windows Vista, such as MCE, Web browser, will see Speech Recognition using the 10 foot experience that is available with MCE for example?
A: Speech works with any application that exposes it's information via standard accessibility techniques. Both IE and MCE do that.
So ... You should be able to use SR in those apps pretty well.
I know that it works very well in IE:
"Start Internet"
"Go to address"
"w w w dot m s n dot c o m"
"Press enter"
"Scroll down"
"News"
Q: [44] I would like to have option of participating in recognition using the context of my application. Will it be possible?
A: Yes. You can use SAPI or System.Speech.
Q: [127] I'd like to congratulate the Speech team on the "show numbers", and related abilities to control windows and programs with speech!
A: I love this feature! It enables the user to accomplish what they need to on a page even if they're not sure what to say. So it helps you get your tasks done.
Q: [84] will i be able to use speech recognition with another keyboard/regional setting that us/english which is awkward and klunky way just to use the speech recognition
A: You can switch the OS to another local by installing a MUI pack and if there is speech recognition available for it you can then use that language. You cannot use a different language than the current OS language using the Speech UI. You can use a recognizer in a language different from the OS (installed with a MUI pack) inside a program that uses the recognizer in process.
Q: [102] Do you recommend a bluetooth headset for it?
A: No particular recommendations, given the limitations of BT headsets currently (see my other answer on BT headsets). However, I'm a fan of the Jabra 250v for my cell phone and as an occasional headset.
Q: [142] will speech recognition be available at logon?
A: Speech Recognition isn't available on the log on screen itself, but once you've logged on, if you chose to start Speech Recognition at start up, it'll be there ready for you to use.
Q: [146] Is Japanese speech recognition supported in the just released Japanese CTP?
A: Japanese speech recognition will be supported in Beta 2
Q: [135] any way to eliminate some of the background noise, or improve the system to recognize your voice over the rest of noise?
A: You can adapt to your change of environments by running the microphone wizard again. You can improve accuracy by taking additional voice trainings. Noise canceling microphones also help remove background noise.
Q: [59] About Q: [8], are you working to support brazilian portuguese?
A: We are not actively working on Brazilian Portuguese for Vista, but we are aware of significant interest for this language.
Q: [82] Can you bring voice training files from XP into Vista? What about training files from Dragon?
A: XP to Vista
Q: [34] Anyone with MS gone to on-line banking and used speec rec to input passwords?
A: Using speech recognition for passwords is more difficult than for other text fields. You'll likely have to use keyboard emulation. Are you having a specific problem, or a issue with a particular site?
Q: [67] will the speach API's be available in Longhorn server in the future, if so could they be used for telephony applications al la MS Speech server ?
A: SAPI and System.Speech will be there. The recognizers won't be installed, though. You'd have to obtain a recognizer separately.
Q: [154] How many custom words can be added to the lexicon, is it an unlimited amount?
A: At this time 1000 words can be added. Do you need more?
Q: [52] Hi, will it be possible to have various languages to select in the final release? I'd like to have all languages available, or at least as free downloads.
A: If you can run your installation of Windows Vista in a particular language that we support, SR will be avaible.
Q: [112] Will there be a differenation between English-US and English-UK for speech?
A: We consider English-US different enough from Enflish-UK to warrant a different recognizer for it. We are actively working on UK-English.
Q: [56] And will the performance be much more increased (or the needed resources be lowered) in the final release?
A: The resources needed in the final release will be very similar to that needed now.
Q: [169] I SR is available for Media Center can this also be used on an XBox 360 and use the 360 headset for control???
A: Not currently.
Q: [111] You mentioned USB headset/mic, does that work better than a mic with soundcard, or did I misunderstand your ans
A: I think *in general* USB headsets or microphones work better, but that's a subjective opinion. It really depends on the quality of the components used. Generally the USB headsets are highly quality than the cheapy headsets that have separate mic and earphone plugs. But if you have a good quality headset and good quality audio hardware, you should get good performance with speech recognition.
Q: [159] which speech app is MS using for Speech Recognition (Learnhaut-Hagtberg??) or dragon you-know!
A: We are using our own speech recognizer that has come out of years of R&D at Microsoft and that is a continuation of the recognition engine that was in Office 2003 for example. We redesigned our own Speech User Interface from the ground up for Vista.
Q: [167] Speech has always been pretty good! How is MS trying to get MORE ppl to utilize it?
A: We're making it better.
More people will use it once it's significantly easier to use, and available in more languages. From a marketing perspective, we haven't finalized our plans yet.
Q: [121] Are there any plans to integrate Text-To-Speech natively in common Windows controls such as text boxes? Such feature already exists in Apple Mac OS.
A: For speech output, Microsoft has included Narrator in all versions of Windows for several years now. It will speak the name of controls as they are selected and/or focused.
Q: [161] Does Speech Recognition support Media Center?
A: In some ways, yes. Although, not in the coolest ways like "Record Survivor at 8 PM on Thursday". Although, you could add a macro to do such a recording if you'd like with a future Macro tool that we're creating.
Q: [133] Will you need to address the voice recognition feature somehow? Or in a different perspective of this, will talking nearby affect or activate the voice recognition feature?
A: User is always in control. Once you have Speech Running you can say 'stop listening' and the Speech recognition application goes into sleep mode. It can be awaken by saying 'Start listening'. If you have a mute button on your microphone, you can also use that to control when the speech application is listening.
Q: [162] Are you going to develop some more advanced API oriented to the processing text from dictation mode (syntax parsing, semantics)? So we will be able to build truly voice interface. And not to make user remember the list of predefined commands
A: I'd like to hear more about what you'd like to see developed. You should send us some mail to follow up offline at listen@microsoft.com.
Q: [155] Who is the main target for speech recog. in vista. In my business, the medical and legal professions have the greatest for this. Is there any special attention given to certain users based on the flavor of the OS ?
A: Vista is shipping with features that could be enable in the future once we get a chance to provide tools to developers to support specialized languages such as medical and law. This is still in planning stage. But the OS will support it. Microsoft does not plan to provide the vocabularies needed. It would likely require ISV's to do so.
Q: [185] Many severely disabled users control the computer entirely by voice. They do not have the option of using the keyboard to type in login credentials. If speech is not available to login, have any alternatives been considered for disabled?
A: What do they use today?
Q: [171] Will Speech in Vista support similar "read back" features in Outlook for PIM info, similar to MS Voice Command on the WinMobile platform? If not, could this be added as a feature request?
A: It's not currently supported, but I'd really like that feature too. Again, a Macro could be created to do this.
Q: [156] Wil SR be available on Longhorn Server, and if so via a Terminal Server session?
A: Speech Recognition doesn't work via Terminal Server currently.
Q: [99] On a somewhat related note, we have customers using the Philips SpeechMike Pro (serial interface) and who want to switch to the USB version. Will they have any problems with compatiblity?
A: They *should* not have problems with the USB versions, but it's highly dependent on the drivers that Philips provides. Today I'm expecting a shipment of microphones from them for testing purposes so I'll find out for myself. Do you know the model number of the mic they plan on using?
Q: [172] Will the user be able to wake up his PC from Standby using Speech?
A: Not at this time. That would be a great feature though.
Q: [82] Can you bring voice training files from XP into Vista? What about training files from Dragon?
A: For XP to Vista upgrade, only user added words will be used for the new profile. Users will have to re-do training. Dragon training files will be transfered on XP to Vista upgrade, but they will not be used by our recognizer. If you are doing clean install than Dragon should have some kind of profile and data transfer tool/system.
Q: [163] My understanding is bluetooth has a lower quality output than other wireless methods, so would not be suitable for voice recognition, is that correct?
A: The Bluetooth standard currently only allows recording in 8kHz (vs. 16kHz typically used for speech recognition on the desktop). This was mainly geared towards mobile phone use scenarios. So we would expect some degradation in accuracy when using a bluetooth microphone due to the lower quality signal.
Q: [186] Do you expect Windows Vista speech recognition accuracy to significantly improve before Vista ships or is what we have now essentially what we should expect in the shipping version of the OS?
A: It will continue to increase, but probably not substantially.
Q: [126] There are several version of Dragon naturally Speaking - one of the main that I have used is the pro- legal and medical what are Vista's speech recognition's capabilities in that region
A: You certainly can write code with speech recognition, but given the amount of symbols, it's tedious at best. I recently read that there are a number of software developers here at Microsoft who use speech recognition on a full or part time basis.
Q: [76] Do you think speech recognition might correct enunciation of some of our teens?
A: Maybe! Correct enunciation will definitively boost the accuracy of dictation.
Q: [69] I wanted to try out the speech functionality, but I've been unable to run Vista Beta 5308. When installing under Virtual PC 2004, I get the error "Windows Setup: setup.exe - Application Error The exception unknown software exception (0xe0000100)".
A: Please log that but if you h aven't already.
Q: [194] Will speech recognition work over remote desktop?
A: Not currently.
Q: [189] main english US version word engines Webster?
A: We use our own lexicon that's been developed for the last 10+ years.
Q: [181] can this be used with external devices to control media center devices, home automation systems, security systems?
A: You could certianly write such a program using SAPI or System.Speech.
Q: [130] Follow up to 71. Are we only seeing US English currenlty in Vista builds?
A: As far as OS languages are concerned, we are releasing German and Japanese as well as English. We will be releasing more languages around beta 2, but aren't ready to announce exactly which ones yet.
Q: [196] Many severely disabled users must have assistants or co workers log in on their behalf. It would interesting to know the position of Microsoft's Accessiblity Division on this issue. What has been their role in the development of speech recognition?
A: We work closely with the Accessibility Technology Group (ATG). Although we're not supporting running Speech Recognition on the log on screen in Vista, it is a goal someday.
Q: [92] Will there be an option for users to export their voice training profiles so they can take it with them to different PCs
A: Yes. You will be able to use "Windows Easy Transfer" under System tools to move speech profiles (voice training data) and other documents and setting from one PC to the other.
Q: [145] Do you know of any 3rd party vendors taking advantage of the newer API to make custom 'answering machine' / 'voicemail' systems with the new functionalities? You could go pretty gonzo if you really wanted to for that type of thing.
A: For telephony scenarios like this one we offer Microsoft Speech Server as a platform and there is a whole partner ecosystem working on various applications in the telephony space. We are also integrating this functionality into Exchange 12 and this will allow you to interact with your emails and voicemails via speech.
Q: [148] Has the introduction of the UMPC plaform influenced changes in the Speech abilities of Vista? It seems like this new platform could really benifit from speech based control.
A: It hasn't influenced it directly, no, but we've been working with the UMPC team for quite some time.
Q: [184] I hope this question is not redundant in this fast and furious chat: Many doctors hope to to use the native VR in Vista. Will that be addressed with more specificity and complexity, with respect to medical vocabularies etc.?
A: The native VR in vista will not do a good job with specialized llanguages such as Medical, Law, Engineering, etc. Medical fields such as radiology and others required specialized vocabularies and/or Dictation Topics (also known as Language models). Vista will be able to use specialized language models other than the one provided natively. What remains is for MS to create the tools needed to enable this feature. This is something we are still at a planning stage. But we are certatinly interested. I would like to hear more from you on this topic.
Q: [110] Will speech recognition detect voices of different people and respond according kind of like individual "voice profiles"? Will I be able to have "voice passwords" that only my voice can execute?
A: Currently the speech profile is keyed to the logged on user. Let's say that Bob is the currently logged on user and he has setup and trained the speech profile and uses it regularly. If his daughter Barbara starts using speech recognition, under the Bob user account, the quality of the recognition will be poor. Also, the speech profile will, over time, attempt to adapt to the new voice and thus make the recognition for Bob poor as well when he uses it. So, it's not recommended to have multiple people using the same profile.
Q: [90] Speech recognition seems to require certain words to be said with an american dialect or say them in an american way, any comments on this?
A: We train our acoustic models for American English with a large speech corpus designed to represent the range of pronunciations found in American English. We provide two features to make SR better for each individual user: 1) Speaker adaptation. The speaker-independent engine shipped with the OS rapidly adapts to an individual's accent, and becomes a speaker-dependent engine. 2) Adding pronunciations. Users can add new words and new pronunciations to the system.
Q: [94] Will their be a Russian speech recognition? I go there a couple times a year on Missions riips and it would be good.
A: Sorry, currently we don't have plans on shipping Russian speech recognition.
Q: [179] Q154: We need to be able to add at least 200,000 words from a medical dictionary to the lexicon.
A: For medical situations, you really need a full blown lanuage model to predict the relative frequency of words occurring in sequences. We are working to make that be available, but that's separate from adding user defined words to the lexicon.
Q: [177] Follow up to 154, is this per user or system total? Per user this figure sounds ok, but total system it may be low.
A: Per user. You can actually add more, but the recognizer stops considering them after the 1st 1000. I'll take this feedback back to the development team and see if there's something we can do to make it better.
Q: [212] How about transferring ViaVoice user data to Vista? Will this be possible?
A: That's no supported.
Q: [107] If I have only a USB mike, how can I rig it to work properly (vs. headset)?
A: Your USB mic should be fine to work with speech recognition. Make sure Vista recognizes it properly and that it is set a the default input device to be used with speech recognition.
Q: [124] Will we have to manually switch between languages or will it auto detect the language being spoken?
A: For the Speech UI in Vista you will only be able to use the languge of the operating system. Applications can run the recognizer in process and can switch between different recognition engines. The application could be designed to switch to another language based on a command for example. There is no explicit language detection functionality as such.
Q: [199] how SR perform www.anypage.com, sr support dictation of uRL's
A: In vista we are improving the way we recognize URL's. Dictating then should work in Vista.
Q: [54] Is this something that might be usefull for the blind? Will there be books read by the home computer and computer commands available with voice commands?
A: Speech recognition is useful for people who need alternative means of input. Some people who are blind also have those needs. Rather, the other tools that Microsoft provides are more suited for visual impairments, such as Narrator and Magnifier. To be aware those, that these tools are *very basic* and are meant to bootstrap users into the platform. There are many companies devoted to providing a much better experience for people who have low vision or are totally blind. See http://microsoft.com/enable/ for more information.
Q: [214] will vista speech be 508 compatible?
A: Yes, it should be.
Q: [204] Will Voice Recognition work the next build? Launching it currently crashes rundll on all of my systems, and yes it's been bugged
A: It should work much better in Beta 2.
Q: [198] Does Speech recognition have a built in translator?
A: No not at this time.
Q: [70] will the speec engine integrate and transcribe from digital voice recorders
A: We won't support that feature out of the box. An application could be written to that though, using SAPI or System.Speech.
Q: [95] What are the languages (besides English, German and Japanese) that will be supported by Speech Recognition; and how can users 'install' definition files for other languages?
A: To install other languages supported by Speech recognition, you will need to install MUI packs and switch OS to that language locale. Than you will be able to start Speech Recogniton for that language.
Q: [79] I am very curious to how speech works with office 12 is it the same with same function as it is in Office 2003
A: Office 12 actually removed all of their native speech support, but when you run Office 12 on Vista, you should get an even better experience than what you had in Office 2003.
Q: [223] Can SR be used to switch between apps?
A: Yes you can use the command "Switch to <application_name>" to switch between running apps.
Q: [68] will speach recognition be integrated with intelasense?
A: Good question RLM, there is nothing specific in the Speech UX for IntelliSense currently. Can you give an example of how you'd like it to work?
Q: [85] Is there a guide to use for trying speech recognition, perhaps a URL, to get us staarted?
A: You could take a look at this screen-cast. It has a good introduction to what you can do with Speech Recognition in Vista. http://blogs.technet.com/chenley/archive/2006/02/21/420136.aspx
Q: [202] Will VOIP be a a possible input source, a la WM5.0 devices?
A: No for the main user experience but you can write an application to do it.
Q: [101] re [46], why would an USB mic be better than a analogue one?
A: Yes. You can also use your analog microphone with USB Pod.
Q: [114] I used dragon naturally speaking professional there is a training session in each program its suppose to be totally hands free , can you train speech recognition to add on new words its not familiar with or is it just a light program
A: We provide a totally hand free tutorial/training session as well as additional voice trainings with Windows Vista. You can also use the speech dictionary feature to add and train new words. We also have a document review feature that scans your documents/emails if you want and picks up the words you frequently use. Are you expecting more features in this area?
Q: [229] is there a total number of words you can add to the vocabulary? Can you add a list (an ASCII listing), of words directly to the vocabulary.
A: 1000 words max, currently. We don't have UI to add words in bulk at the moment, but it's a good feature suggestion. An applictation could easily be written by a 3rd party to do this.
Q: [221] do you think voice recognition could help you to get that data and translate it to other languages??
A: Translation is not a feature in Vista. It is conceivable though that a 3rd party could use the speech recognition capability in Vista for one of the supported languages as an input to their translation software.
Q: [85] Is there a guide to use for trying speech recognition, perhaps a URL, to get us staarted?
A: Try this:
http://www.microsoft.com/technet/windowsvista/library/c208e792-e591-455a-82d9-a98264324e0d.mspx
and this:
http://blogs.technet.com/chenley/archive/2006/02/21/420136.aspx
Q: [213] Will very specific professional terms of VR dictation supported in the future? For example spino-cerebeller degeneration in medical?
A: Vista native dictation support is for standard English more like the one you find in news papers for the general public and business community. For medical, the system requires a specialize dictation topic to be created and installed. Vista will support such thing but MS would need to provide the tools required to build them. We are still at a planinig stage for this. We would appreciate your feedback
Q: [128] This is not exactly about speech recognition, but also a speech-like feature: Why does the "Narrator" have a female voice now? I liked the male voice (which was called Sam in 2000/XP) better.
A: The reason is that we changed the TTS engines in Vista to reflect new technology. Microsoft Anna is a step above what was previouly available. I cannot say with certainty that a male counterpart will ship in the final release of Vista.
Q: [236] re [101] (and [46]) are there any technical reasons for why a usb mic (digital) would be better than an analogue one? If that's the case does that not indicate that your A/D conversion suck?
A: The A/D conversion is done in hardware. Thus, because AC97 hardware is on the motherboad, there is sometimes electrical noise introduced. With USB, the A/D is done outboard and thus less prone to that noise.
Q: [83] Is there a point to use speech in a single room office with many people around?
A: Three things come to mind here: 1) The main challenge of that scenario is capturing sufficient quality audio from the multiple speakers - quality microphones and setup to provide a reasonable signal to the SR. 2) The speech recognizer adapts to each user to improve recognition accuracy, and links the adaptation information to that user's profile. A profile for one user may or may not work as well for another user. 3) Multiple speakers on a single audio feed produce overlapping waveforms. SR is designed for a single speech stream. I can envision a future when multiple parties in a meeting could each have their own mobile device capturing their speech (speaker-adapted recognition, good quality audio input), which is then synched up with the rest of the meeting!
Q: [187] Will the system be able to read back ASCII books, interpreting punctuation correctly?
A: Narrator and our TTS engines do a excellent job of speaking even complex phrases
Q: [116] What apps will natively work with Vista speech recognition; Office, OneNote, IE, others?
A: Most applications will work with speech recognition!
Q: [131] Will speech also work in a way, that you maybe start a program by speech command the program is reporting back via speech output "Program xyz is ready now"?
A: You could write a macro that does this.
Q: [136] Will there be a Vista Logo Programme for headsets for SR?
A: Not specifically for speech recognition, but the Designed for Windows logo program is available for hardware. You can be sure that if a device has the Designed for Windows Vista logo, it will work with Speech.
Q: [224] Have you guys made any progress towards having multiple SR applications be able to share the same feedback and commands windows? Or does each application that uses SR have to provide its own?
A: Hi Mike. The feedback window is programmable. You can send your text feedback there. The What Can I Say window is still not configurable by ISVs.
Q: [237] I've noticed that sometimes when I play music on my computer loud enough, SR picks up the noise. Is SR going to be able to block out the noise coming from that same computer?
A: Doing echo cancellation to remove music playing from the computer in the background is not a feature at this point. To get the best recognition accuracy I would recommend that you either mute or at least turn down the music.
Q: [225] Scenario: I have a MUI system and use German as my default language, train some German words. Then I switch to a different language, say Spanish - will my training and my custom words be remembered when I switch back to German?
A: Yes they will
Q: [230] como configuro o windows vista para usar o memory key para extender a memoria ram?
A: Hola esta pregunta no es acerca de Speech Reconigtion. I can't answer
Q: [218] SR in vista is great, it's one of the best SR's I've used, what are the plans to "advertise" sr? As in make the average user know it's available.
A: Thanks. Please tell us what you like at listen@microsoft.com The Windows Vista marketing team is working on that. We have a web presence already here: http://www.microsoft.com/windowsvista/features/foreveryone/speech.mspx
Q: [105] What's the recommended course of action when implementing a command in an application that conflicts with the systems built in commands?
A: Vista's speech APIs have a prioritization mechanism. The application can choose to to override commands, or some commands might be overriden by the SR itself.
Beyond that, it depends on the command. If you want to follow up, feel free to send email to listen@microsoft.com and we can discuss it offline.
A: We have similar accuracy.
Q: [97] Suggestion: What about using speech as a password/log on in the future? It will detect the voices of us as an alternative to typing our password
A: Jabez, that is an excellent suggestion and it's something that our security people would love to see happen as part of a multi-level authenication scheme. However, given the challenges of just getting it right on the desktop, it'll be awhile before we can "speech enable" the login process.
Q: [143] you say you "will look at your documents and emails and add words we don't know to the lexicon". Will the user have some control over that? Which documents to harvest? Which harvested words to accept?
A: We use the same system that the indexer in Vista uses. So, if you can use Search to find the document, Speech will consider it when it updates the language models.
Q: [139] Will there be any special add-ons for the speech dictionary to include things like special medical and legal terms?
A: Vista native dictation support is for standard English more like the one you find in newspapers for the general public and business community. For medical, the system requires a specialize dictation topic to be created and installed. Vista will support such thing but MS would need to provide the tools required to build them. We are still at a planinig stage for this. We would appreciate your feedback
Q: [176] will speech be smart enough to "stop listening" if i get a call on a softphone or voice chat?
A: Not currently, but that's a great suggestion.
Q: [9] Is a noise cancelling mic necessary for good results?
A: A good quality mic will help with recognition accuracy. Noise cancellation might help in somewhat noisy environments, but probably won't make a noticable difference if you use it in a quite environment.
Q: [178] When developing applications that make use of speech recognition in Vista, will the software logo requirement require System.Speech, or may SAPI be used as well?
A: You can use either one.
Q: [158] In an Enterprise environment, will we be able to control how much control or who gets access to speech?
A: Yes. You could lock down the registry location we configure SR to automatically start.
Q: [174] Say please, probability is which that there will be support of Russian in Beta2.
A: We are not actively working on Russian for Beta 2.
Q: [205] I have used the speech typing tool in XP, it took a really long time to get it set up, Will Vista be the same? Is there a way to import the file from XP to Vista?
A: No, the training process is more straight forward, and is intergrated with a tutorial on how to use the system. No, you cannot import the speech profile from a previous version - even if you could, it wouldn't help much - the recognition in Vista is much better even without a trained profile.
Q: [245] Do I understand correctly that you can NOT currently use voice to "open word" "save" "print" "close word" and shut down the computer, stuff like that? Is that something that is coming?
A: We DO support in Vista the commands you are mentioning above. It is already in the product.
Q: [242] could you know the msagent team ?
A: We also own MSAgent.
Q: [149] How well will speech integrate with other Microsoft and 3rd Party applications without to much user hassles?
A: There are two level of API's one for COM based applications (SAPI) and one for Managed code. If you have a specif quetsion on what exactly you want to integrate please ask. Thanks.
A: I am not sure about your question, but USB microphones are OK. The difference will be USB vs. Analog, and we recomend USB.
Q: [246] Will there be a "push button to listen" option where voice will only be listening when I hold a button down?
A: Sort of. "Ctrl-Windows" turns speech on and off. Would you like to see push to talk added? What about configuring the key?
Q: [231] Are there any serach specific special commands exposed in Vista SR?
A: No but you can control the search UI via voice.
Q: [144] Will lexicon expansions be made available to the OS via Windows Update?
A: Yes, SAPI allows for two levels of lexicons, Application and User level lexicon
Q: [245] Do I understand correctly that you can NOT currently use voice to "open word" "save" "print" "close word" and shut down the computer, stuff like that? Is that something that is coming?
A: You CAN currently do all those things with Speech. You should be able to say any command (menu item, dialog box, etc.) you can see on the screen. Rob Chambers has a bunch of commands listed in his blog.
Q: [228] Narrator crashes occasionaly, whould that affect speech input?
A: It should not affect speech input. Please, please, please file bugs against any crashes
Q: [244] Is microsoft working with computer/monitor manufacters so that microphones will be built into the hardware?
A: Actually, built in mics in monitors and laptops usually do poorly with speech recognition. A high quality microphone is your best bet.
Q: [100] Could you compare accuracy of Vista ASR and viavoice, please?
A: Accuracy is similar
Q: [87] [14] Lesley: Sure your colleagues can use English SPeech recognition, since they all are living in the States! What about your colleagues working from Japan/China with no american accent?
A: True, but they actually demonstrate a wide range of American English influence in their phonologies. You are right to point out that ideally, models would exist for varieties of English beyond US and UK, such as Singapore English, etc.