Related
I have spent some time looking at various shareware and commercial voice dialer apps this week.
I was really impressed with the specification of MS VoiceCommand No training required, context dependent actions (ie you say "Call So and So" and it will dial immediately if there is only one number, or will say "Mobile or Work?" if there is ambiguity, or you can just say"Call So and So at Work" and off it goes. Terrific.
Except it doesn't work on the XDA. Can't see why - it hears something (so "Show Calendar", "Help", "Yes" etc. work) but does not have enough mic sensitivity to do name recognition. This is so whether you hold the XDA 1 inch or 18 inches from your mouth, whether AGC is on or off.
So for a quarter of the cost I have bought NeuVoice Dialer, which works pretty well. Problems? Needs to be trained, no context dependency, so you either have to associate a specific number (Work, Mobile etc) with a voice tag, or you have to recognise the name first, the specific tag second, which means pressing the action button twice. Nause. Also not very easy in the car w/o a hands-free profile, which is what we are all after, isn't it?
So are my experiences unique? Would love to hear!
I'm using Voice Command with XDA II and it works great. How many processes are you running? I found that if you get close to the 32 process max then it doesnt work as well (see other posts in this forum about max processes). At 26 processes no problem.
Only two or three!! It is really strange - I can't see why it works so much worse than NeuVoice. Not noticed any particular probs with the phone either , so I don't think that the microphone itself is faulty, although it could be that.
How well is VOice Command working really? How many contacts do you have and does it really discrimate? I have about 500 contacts+.
Cheers
Stephen,
I doubt you only have 2 or 3 processes running, you mean applications visible through the memory option in settings. You need a third party tool like Taskmanager (available on Handango) to see the processes running (a bit like the Windows taskmanager) I have 500+ contacts and Voice Command really does a great job of discriminating between them. Try the stereo headset that comes in the box to rule out any microsphone issues. Also I run Voice Command with default settings, I didnt mess with the AGC Microsphone settings at all.
Out of the programs I tried for VoiceDialing, I found Fonix VoiceDial was the most accurate (using XDA 1)
Tried with the stereo lead - very, very slightly better, but no way is this stuff usable.
I'm pretty sure there is an architecture issue here - the fact that the AGC cannot be manually set, strange behaviour using teh recorder during phone calls etc. VoiceCommand may be great, but it don't work too well on my XDA2. Maybe I have a funny voice!!!
Cheers
XDAII - MS Voice Command experience
I am using MS VC with i-MATE for already several months and I have over 700 contacts...
Based on my experience - try to reinstall MS Voice Command - some times it works better.... Also do not speak to loud; do not keep it very close to your mouth. It works fine even if it is on a console between sits and I am driving (looking straight on the road).
And yes, you should check number of process running... http://www.scarybearsoftware.com/ppc_cn_overview.html - would be the best bet...
Good luck.
Paul
stephen_oliver said:
I'm pretty sure there is an architecture issue here - the fact that the AGC cannot be manually set, strange behaviour using teh recorder during phone calls etc. VoiceCommand may be great, but it don't work too well on my XDA2. Maybe I have a funny voice!!!
Click to expand...
Click to collapse
Don't forget, you need to speak with an American accent it you want this thing understand you...!
Yes, that has improved it - 135 notifications.
Not sure how to check active processes - got any suggestions on utilities?
I can now get recognition about 60-70% of the time. Sadly it still has problems with my "Yes"!! Maybe I should coach my voice to sound more patient :?
Thanks for the advice ...
How do you BUY VoiceCommand?? Won;t accept non-US addresses!!!!
As far as I know you can get trial for one or several days.
As per "buy it" I am not sure how you can do that not from states or Canada.
About YES - possibility that you just saying it too soon. Try pause for 1,5 2 seconds and after repeat if necessary.
Thanks - I've discovered Handandgo/uk so I'm all set now.
Cheers everyone!
Hate to say this but I tried Voice Command and gave up!
Sometimes it got it right, but for most of the time it got it wrong. I think I had better results when I tried an American Accent with a Scouse Accent (Livererpudlian for our international fans), but even then it was a burger less than a MacDonalds.
I seemed to think it was down to the XDA II only having AGC on and off choices and not a manual audio gain setting. I did try it with both the free version and the *cough* cracked version (for evaluation purposes of course) to little success. Saying that, didn't have much success with Via Voice either.
Anybody willing to "share" this software.... long term.. hehehehe
e-mail me at [email protected]
fella's please persit with and you will be rewarded with porbably the best voice recognition program available....
i nearly gave up but i'am glad i did'nt
It helped me going to the website and reading the help files and watching the demostration video...
Zetex
MS Voice Command and Fonix
G'day Guys,
I've persisted with MS Voice Command and agree with the previous posts.. If you've got your alerts sorted out and processes running low, it's a REALLY good package...
The Fonix VoiceDial and Voice Commander has excellent potential but I couldn't get the accuracy up and the volume was lousy - Crackly and distorted.. So I'm sticking with MS VC...
Still can't get "DIAL 1 2 3 4" etc. working AT ALL.. It's ALWAYS wrong?!?! But "Call CONTACT NAME" is consistently excellent.. So that's all I need it for!
Beats the heck out of having accidents on the road while trying to stylus the iMate!!!!
P.S. For the SmartPhone, Fonix VoiceDial is absolutely brilliant - I've also used that and it's consistently good.. Just a shame that there seems to be an issue with the iMate PDA (O2 XDA II) hardware and Fonix VoiceDial...
Cheers
Andy
Voice Command in Oz
Andy,
How did you get VC in Oz?. I have used the trial version and found it great (after learning to americanise my "strine"), but non US addresses just can't download. I am happy to pay for my programs, but if MS don't want my money...
Che?
Voice Command in Oz
Andy,
How did you get VC in Oz?. I have used the trial version and found it great (after learning to americanise my "strine"), but non US addresses just can't download. I am happy to pay for my programs, but if MS don't want my money...
Che?
usa purchase
Its very simple if you are only purchasing a software download - Give your real email address but gìve any usa mail address that you can see on packaging or from the web and the sale will go through! important the ZIP Code must be real
Neuvoice for XDA II
Thanks for that Sam. In the meantime, I have tried Neuvoice, and have actually found it is a better product. Apart from the need to train it, the recognition is better and I don't have to alter my accent
I've started developing an application that allows the user to compose and send an email completely hands-free... by voice command only.
However I'm having trouble finding a decent, open source (free) speech recognition (speech-to-text) engine / API to use.
Does anyone know of one? I tried PocketSphinx but had trouble compiling it in Windows using VS2008.
I'm wondering what API the Windows Live Search app uses? Its speech recognition capabilities are already decent, and if it's included with Windows Mobile or .NET Compact Framework 3.5 or Windows Mobile 6.1 itself, then I would prefer to use that. But I'm having trouble determining if this speech recognition is available to 3rd-party developers and, if so, how to interface with it.
Any help would be greatly appreciated!
OMG I hate timeouts lol
So I had this nice long post about how I thought it might be one of three things and I whipped out my omnia and disconnected the network and blah blah.
When I hit post, I got a not logged in timeout.
So here's the short of it:
It uses a server, that's probably related to UC aka Office Communications Server aka Speech Server 2007... you can get to it (and all the Microsoft Speech technologies, including Voice Command) here:
http://www.microsoft.com/speech/speech2007/default.mspx
A little more searching lead me to read the MSDN Channel 9 blog on said subject:
http://blogs.msdn.com/speech/archiv...h-for-mobile-now-with-speech-recognition.aspx
which states:
"The speech recognition functionality for the application doesn't actually sit on the Windows Mobile phone. Instead, the phone takes your speech input, sends it to a server, the server does it's recognition magic, and sends the results back to the phone. "
Speech Server 2007
Thanks for the reply MerlinJim... sucks about the timeout! That's why on a long post I always copy the text to the clipboard... that way if it times out I can just paste it in! (It's happened to me too many times for me to not do that now!)
Yeah I've looked at Speech Server 2007 as well... and I was thinking that maybe Live Search offloaded the speech recognition to a server. There's a little lag between what you say and when it guesses what you said.
I guess something like that would work. If you're writing an email then you need an Internet connection, and so sending the voice data to a speech server would be plausible. The only downside would be if it used up a lot of data transfer/bandwidth, and the user was on metered bandwidth.
The lag would be a bit of a drawback, because if the Speech Server guessed incorrectly what you said, but you kept talking (due to the processing lag), then you would have to go back and correct what you had said.
And also sometimes the Live Maps speech recognition is WAY off. Like I'll say "1 Jefferson Parkway" and it will come back with something like "Did you say 'Parkstone Apartments?'"
It's also speaker-independent, so you don't do any training. I would rather train an app to recognize my voice specifically, because I would be the only user of it.
But it may be my only solution for right now. Thanks for the info! I was beginning to think that no one knew the answer.
acrosser said:
Thanks for the reply MerlinJim... sucks about the timeout! That's why on a long post I always copy the text to the clipboard... that way if it times out I can just paste it in! (It's happened to me too many times for me to not do that now!)
Yeah I've looked at Speech Server 2007 as well... and I was thinking that maybe Live Search offloaded the speech recognition to a server. There's a little lag between what you say and when it guesses what you said.
I guess something like that would work. If you're writing an email then you need an Internet connection, and so sending the voice data to a speech server would be plausible. The only downside would be if it used up a lot of data transfer/bandwidth, and the user was on metered bandwidth.
The lag would be a bit of a drawback, because if the Speech Server guessed incorrectly what you said, but you kept talking (due to the processing lag), then you would have to go back and correct what you had said.
And also sometimes the Live Maps speech recognition is WAY off. Like I'll say "1 Jefferson Parkway" and it will come back with something like "Did you say 'Parkstone Apartments?'"
It's also speaker-independent, so you don't do any training. I would rather train an app to recognize my voice specifically, because I would be the only user of it.
But it may be my only solution for right now. Thanks for the info! I was beginning to think that no one knew the answer.
Click to expand...
Click to collapse
perhaps, but there IS a speech application loaded ON a Windows Mobile 6.1 which has text-to-speech capabilities and speech recognition
(my Blackjack II loaded with Wm6.1 has this capability)
can't find any API to use it though... only way to activate this TTS capability is to
1) sms announcing
2) appointment announcing
3) call announcing
no actual program to do TTS...
Any progress on this or any other speech-to-text program? I'm really interested in finding one.
Wouldn't mind being a beta tester, either.
*Double Post*
DELETE
DIRECT LINK for the CAB
MS release a technical preview of recite:
What Is Microsoft Recite?
Microsoft Recite is a search technology for your voice that runs on Windows Mobile* devices. With Microsoft Recite, you can use your voice to easily store, search and retrieve the things you want to remember, where and when you need them. Microsoft Recite is available as a free technology preview beginning February 16, 2009.
*Microsoft Recite can be used on devices running Windows Mobile version 6.0 or higher. Not sure what you’re running? A complete list of devices can be found at http://recite.microsoft.com
How Does It Work?
Microsoft Recite’s voice search makes it easy to retrieve your stored thoughts and notes by using voice pattern matching. It analyzes the patterns in your speech and finds matches between two recordings -- the notes you stored on your phone, and the search you do using your voice. With Recite you can store thousands of spoken notes, and then later retrieve the notes you want based on a match with your search term(s). This is different from speech recognition, which has to accurately convert spoken words to application-readable input.
Press “Remember” to record a thought.
Press “Search” to retrieve your thoughts. It’s that simple!
Consumer Use
We can think of countless handy ways that you might use Microsoft Recite… record your shopping list, friends’ birthdays, addresses, school happenings, gift ideas, get togethers, favorite wines… anything you might need or want to remember later. Recite even lets you remember and search in multiple languages.
Here’s an example. Imagine your co-worker, Paul Johnson, tells you about a book that he thinks you might like, Hot, Flat and Crowded, by Thomas Friedman. To start recording a mental note, launch the technology, press the 'Remember' button, then say what you would like to record; in this case, “Book recommendation from Paul Johnson: Hot, Flat and Crowded.” Next, press the “Finished' button to complete the recording and store the note. Later, when you’re ready to buy the book but are unsure of the title, click on the ‘Search’ button and say what you would like to recall. In this case, you might say “book recommendation,” then press ‘Finished’ to begin the search. Recite will then retrieve and play the book recommendation for you.
Or, you might recall that Paul told you something that you wanted to remember, but forgot what it was. In this case, click ‘Search’ and say “Paul Johnson.” Microsoft Recite will retrieve all mental notes that include the sounds “Paul Johnson.” In this case you would hear “Book recommendation from Paul Johnson: Hot, Flat and Crowded.”
No search button - solved, but not capturing voice
I have "remember" on the left softkey and "privacy" on the right one, which loads PIE at the privacy policy page for the project. I can't locate the search funtion. Anybody get this to work?
I have similar funtionality through Evernote, but not the ability to voice search, so it would be interesting to try.
edit: It was not capturing the recordings, so there was nothing to search. Once I got it to successfully capture, the search function was there. It is still shutting down without capturing.
It crash on many HTC .. so its cool but buggy
It doesn't seem to work for me on WM 6.1 (Touch). I can record lots of things but the search always comes up with the same result even if NONE of the words in it match.
No good so far.
New version out. Much improved
I had reported my issues on the feedback page on the beta site, and got an email response last night that a new version is available. I dl'd it and it is much better than the original. It now responds to the touch screen as well as the soft keys, and it captured the recording right 4 out of 4 tries. the search funtion worked pretty well, but not perfect. Saying "Anthony" didn't work, but "Anthony's" did. As in "Anthony's birthday is . . . "
What this program needs to rock is text to speech, and integration with outlook. the ability to speak calendar entries and to-dos into the phone would send MS to the top of a lot of peoples lists.
Yeah this is a cool program. Thanks for letting us know there was an update.
It´s working, but bull.. sh..t
you can´t organize the files you recorded . A delete function is missing .
MS is trying to tease us.
It worked well in my Prophet. But one thing worries me: where are the recorded files saved? It seems internal memory but where?
Could anyone post the Last cab of recite ?
ok got it!
It need to use ie mobile
for downloading the cab !
MS dont like opera!!
MS recite CAB for everyone
I've uploaded the last version that run fine.
Works perfectly on Elfin
Updated again..here it is.
I just found this program and for some reason I am unable to dl it from the microsoft website. Is the cab right above this one the most recent?
Greetings,
For reasons similar to an earlier poster, I'd like to find a way to map a button to mute/unmute the mic during a cell call.
A small application was posted here: http://forum.xda-developers.com/showthread.php?t=386810 which does this.
However it does not work on the Treo800w (as well as a few others) due to the fact that it looks for text on the button, which isn't there in this model.
In any event, after exhaustive searching - their doesn't appear to be any ready-made solution for this. I'm willing to write my own, but it's not even clear to me from the msdn api docs that it's even possible to directly manipulate this? Is this type of control "protected" from the Windows portion of the phone?
The idea would be to write a tiny app to toggle it and then map it to aebutton or something similar.
This is pretty much exactly what "itsallwright" did, and I'd like it for the same reasons - it's just his approach doesn't work on the Treo800w.
Thoughts?
-C
I have a rather interesting idea, and I thought to post it here. I apologize if there is a more appropriate location for this thread, but, as I am new to XDA-Dev, I hope this will be forgiven, and someone could perhaps correct my mistake and move it or tell me what the proper thread location would be.
Also, as a recently-graduated high school student, I have plenty of free time before I start college, wherein I plan on learning a few programming languages, starting with Java, so I can develop for Android and iOS. This means that here at XDA-Dev, I'll hopefully soon become a very active user.
Now that the formalities are out of the way, I'll move on to my idea.
My phone is a Samsung Replenish(SPH-M580), which is a 600Mhz ARMv6, with 512M of RAM, EV-DO Sprint data network that saves me $10 monthly(Why I own it). The drawback to this phone is that even though it has multi-touch, the screen is a tiny 240*320 pixels. Now, I love Android, due to how open it is, but I have an Android-based phone and an iPod Touch 3rd Generation for my music needs(32GB).
When I'm at home, I don't touch my phone near as often as I do my iPod, and when I'm sitting on the couch browsing the web on my iPod, I don't want to have to keep switching to my phone. So my idea is this(And please, tell me if something has been made): If you can have replacement SMS clients such as Handcent and GO, then it should be possible to have an app on your phone that gives you access to your text messages in a browser-- But not just desktop PC browsers(Such as the access given by WebKey or PDA-net), in other mobile browsers, such as Opera Mobile, mobile Safari, etc.
I know it is possible, given the ease of using something like Handcent SMS, and I'm sure it would need root elevation unless there's a good userspace web server, but the question is, has it been done? If not, I'd like to try to get some help in the community with making it.
Sorry for the horrendously long post, but I think it's a good idea that should be attempted, at least.
As a Sprint user
I am also a Sprint user and I'd like to inform you of Sprint's great deal with Google Voice. (As a new user I can't post links so try Google for "Google Voice")
This effectively gives all Sprint users the ability to text via Web or any device with a Google Voice app at no additional cost.
I doubt that it's optimized for mobile browsers of small devices, but MightyText does work on many browsers. If you're ambitious enough, I'm sure you could write a script to optimize the layout yourself for a small device like an iPod. You might want to try downloading the MightyText app (listed as "Free Texting <-> PC SMS Messaging" on the market for some reason) and then going to "mightytext.net/app" on your iPod and seeing how it looks.
Alright, I tried out MightyText, and in the Mobile Safari app, MightyText looks like a scaled down desktop website. It would be cool if they added a Mobile view as well, but so far, it doesn't seem so.
mike12489 said:
I doubt that it's optimized for mobile browsers of small devices, but MightyText does work on many browsers. If you're ambitious enough, I'm sure you could write a script to optimize the layout yourself for a small device like an iPod. You might want to try downloading the MightyText app (listed as "Free Texting <-> PC SMS Messaging" on the market for some reason) and then going to "mightytext.net/app" on your iPod and seeing how it looks.
Click to expand...
Click to collapse
The mobile site for Google Voice is optimized for Mobile viewing. Also the Google Voice app is much better than MightyText as it integrates very tightly with your phone number.
legacyweapon said:
The mobile site for Google Voice is optimized for Mobile viewing. Also the Google Voice app is much better than MightyText as it integrates very tightly with your phone number.
Click to expand...
Click to collapse
I don't like the Google Voice app's way of doing things. I don't like that there's not an option to disable notifications from the Google Voice app itself.