Here is a great feature from Windows 7, “Speech Recognition”. The computer will do what ever you say to it. It will be a great feature to the people who hate to use the keyboard or mouse. Here is how to configure the “Speech Recognition”?

Read More……

Recently I saw a interesting tweet from Seth Sandler regarding a AS3 speech recognition lib. The demo was very impressive. So I wondered why not for python ?. So I started looking around for some premade librarys in Python. I found three good ones

• pyspeech
• dragonfly
• Sphinx from CMU

The first two where windows only solution so they where rejected immediately. The Sphinx library from CMU is very vast library. After some reading I found out that the dwarf version of Sphinx called pocketsphinx is just a plugin for Gstreamer. It was very easy to integrate them and finally integrated them to PyMT. . I will be posting the example app as well as Speech Recognition engine for PyMT will be included in the 0.4 release. Can’t wait to see how users will use this new capabilities in their apps

Other then this i’ve been working on completely rebuilding the animation framework for PyMT 0.4 and its complete now WITH the wiki documentation. Thanks Mathieu for all the advices you gave me in improving this. Heres a video of the app he made with the animation framework.

This one I did.

We are trying our best to release a Alpha version before Christmas, lets see how it goes. Lots and lots of changes and feature additions

Although I’m able to turn pages and make good use of my BookChair, there are times that I would rather read on the computer. This is mainly because of neck trouble and trying to learn C++; getting the keyboard out of the way, setting the book on the keyboard table, turning the pages and hoping Dragon doesn’t pick that up, and glancing up and down from the book to the screen make it a little more cumbersome than I would like. I’ve been thinking it would be nice to just vocally switch windows; read the book on the screen, then go into Win32Pad to dictate some code.

It turns out that I can potentially do that now using Amazon’s Kindle for PC application, which is free (with the exception of the books, of course). I hadn’t had very high hopes for the application’s accessibility, considering that Amazon isn’t great with it (see the lack of captioning for Unbox video products). However, it’s actually not that bad. (Better with Vocola commands, which I will create shortly.) While all buttons must be accessed by mouse commands, actually reading isn’t a bad experience at all.

C++ without Fear was not available for Kindle, but C++ Primer Plus was, so I downloaded a free sample to test how Kindle PC would work with Dragon. To open a book, mousegrid to the Home button (yay, verbing nouns) and click it if it’s not already highlighted. Then, by the same method, double click the book icon. The clicking isn’t as tedious as it sounds because the buttons are large; the grid doesn’t require much narrowing. To turn pages, say “Go left” [previous page] or “go right” [next page]. Up and down do the same thing respectively. To turn multiple pages, say “Go <direction> <number>.”

Text is very clear, and there is a button to adjust the font size. There are also a Go To and a Bookmark button, but Kindle PC already saves your last position when you exit, so that you can open the book where you left off. Note that I didn’t say page number — Kindle doesn’t use them. That somewhat disorients me, but for something like a programming text I suppose it would keep me from getting overwhelmed. That is, seeing 20% read looks more encouraging than “Oh god, I have 900 pages left. Dammit I’m slow!”.

At the moment you can’t create notes, unfortunately. (Not that Kindle would be select and say compatible even if you could, but I’m hoping for something that would allow at least partial dictation, a la Foxit. You also can’t copy text, for understandable reasons, but that sucks for a programming book. I do wish there could be a PDF style layout option, by which I mean the possibility of an autoscroll feature. Kindle PC is nowhere near perfect, but it is passable.

VoiceXML is designed for creating audio dialogues that feature synthesized speech, digitized audio, recognition of spoken and DTMF key input, recording of spoken input, telephony, and mixed initiative conversations. Its major goal is to bring the advantages of web-based development and content delivery to interactive voice response applications.

VoiceXML (VXML) is the W3C’s standard XML format for specifying interactive voice dialogues between a human and a computer. It allows voice applications to be developed and deployed in an analogous way to HTML for visual applications. Just as HTML documents are interpreted by a visual web browser, VoiceXML documents are interpreted by a voice browser. A common architecture is to deploy banks of voice browsers attached to the Public Switched Telephone Network (PSTN) so that users can use a telephone to interact with voice applications.

While HTML assumes a graphical web browser with display, keyboard, and mouse, VoiceXML assumes a voice browser with audio output, audio input, and keypad input. Audio input is handled by the voice browser’s speech recognizer. Audio output consists both of recordings and speech synthesized by the voice browser’s text-to-speech system.

More info from http://www.vxml.org and http://www.voicexml.org

VoiceXML example

VoiceXML has the following advantages

• Leverage of the World-Wide Web and of its capabilities.
• Latest developments in computer-based speech recognition and text-to-speech synthesis.
• Spread of Internet to everyday life

Another important mention is Automatic speech recognition (ASR)systems .

Over the phone, and with no speaker training, a speech recognition system needs to be given a set of speech grammars that tell it what words and phrases it should expect. Recognition over mobile phones in noisy environments, while problematic, can be improved with a new technology called distributed speech recognition, where the early analysis is done on the handset. Advances are also being made in speech synthesis, or text-to-speech (TTS). They can be hard to listen to, and at times even incomprehensible. But newer TTS systems are much more lifelike – they use a technique called waveform concatenation, in which speech is generated from libraries of pre-recorded waveforms.

Why is VoiceXML important ?

First, the phone is important. There are over 1.5 billion phones in use, far more than there are Internet-connected computers. Phones are easy to use and don’t need to be booted up. Telephone networks are much more reliable than data networks.

Second, voice is important on the phone. Voice has always been the natural mode of communication for phones. Even though some mobiles have WAP/XHTML browsers, their small screens and keypads make micro browsers hard to use, especially while driving. The i-mode system is more compelling, though shares the same limitations.

Automatic speech recognition or computer speech recognition converts spoken words to text. The term “voice recognition” is sometimes used to refer to speech recognition where the recognition system is trained to a particular speaker – as is the case for most desktop recognition software, hence there is an element of speaker recognition, which attempts to identify the person speaking, to better recognize what is being said. Speech recognition is a broad term which means it can recognize almost anybody’s speech – such as a call-centre system designed to recognize many voices. Voice recognition is a system trained to a particular user, where it recognizes their speech based on their unique vocal sound.

Both acoustic modeling and language modeling are important parts of modern statistically-based speech recognition algorithms.  Modern general-purpose speech recognition systems are generally based on Hidden Markov Models (HMM). These are statistical models which output a sequence of symbols or quantities. One possible reason why HMMs are used in speech recognition is that a speech signal could be viewed as a piecewise stationary signal or a short-time stationary signal. That is, one could assume in a short-time in the range of 10 milliseconds, speech could be approximated as a stationary process. Speech could thus be thought of as a Markov model for many stochastic processes.

If you feel like developing your own application download the CSU Toolkit. It is a comprehensive suite of tools to enable exploration , learning and research into speech and human-computer interaction.

Applications best suited for VoiceXML

Information retrieval(IR) is a good example for VoiceXML. In such an application the audio output tends to be a pre recorded infromation. Voice input can be a few browsing commands or a whole street address , it can be simple as well as quite rich. Let us consider using this IR application for weather information , where people can call a number and use post code as a commands to retrieve the weather forecasts. Unified Messaging is another example where emails can be read over the phone , outgoing email can be recorded . Directory assistance applications also work well with VoiceXML, this is already demonstrated by many telecommunication providers.

What is the future for Voice XML

The W3C is trying to ensure a VoiceXML standard is available to ensure interoperability between different implementations of VoiceXML can execute the same content the same way. There could be standardised audio playback controls . Speaker verification is a major security concern that needs to be addressed to ensure the authenticity of the messages. Support for multimodal markup standards is also a proposed solution.

It was only after I published my last post that I realized there was something very different about the font – which is supposed to be fixed in this WP template.  So how did it get so humungous?  No, WP isn’t to blame –  it was the Windows Speech Recognition program on my new Lenovo  G550 BAQ laptop.  I had been breaking it in, training the computer (that makes it sound like some kind of wild beast, don’t it?) by dictating to it -  and I had what I thought was a bright idea… I’d dictate the post!! I had a niggling doubt whether it would work on the WP platform – perhaps it was configured only to MS Word?  It worked alright – but with a bloated font as you can see!!

Raw unedited speech recognition looks like garbled gibberish – no kidding!  The Help section assures me that the software program gets better with practice.  Until then, Speech Recognition  can be considered the twin of the perverse Spell Checker – you need a firm hand and agile brain to rein both in.  Here is a short verse (that could have well been produced by Speech Recognition  I reckon).

OWED TO A SPELLING CHECKER 

I have a spelling checker

It came with my pea sea

It plain lee marks four my revue

Mix steaks eye kin not sea.

Eye have run this poem threw it

I’m shore your pleased to no

Its letter perfect awl the weigh

My checker told me sew.

           – Sauce Unknown

After  a bit of research on the Net, I found the original version of  the poem: Candidate for a Pullet Surprise by Jerry Zar.

© Sosha Srinivasan

I am happy to report that nothing really got broken in trying to use Dragon NaturallySpeaking 10 with Mozilla Thunderbird 3 RC 2. The toolbar buttons and menus are still accessible, and my custom buttons are still there, though I had to put them back again. Dragon’s Thunderbird 2 commands, like “next/previous message” and “write mail,” still work. Don’t be alarmed if the File, etc. menu is grayed out — you can still access it, as well as the submenus.

Please note, however, that you will need to use the Mouse Grid commands and Tab and Enter to get around the box and buttons for Tools — Options. They seem to have gone the route of Firefox 3, that way. Oh well.

The newest thing about it is the tabbed interface. When you highlight a message and click Read (if you have the button), your message will only open in a new window. But if you say “press Enter,” Thunderbird will open the message in a new tab. As with Firefox, you can speak a word in the tab — like “Inbox” — and Dragon will click the tab. EDIT: I may have spoken too soon. The only tab that Dragon will click is the 1st one, the main one — Inbox, or Sent, or whatever folder your messages are in. To navigate around any other tabs, you will have to use the “press Control <number>” command.

The “close tab” command doesn’t transfer, because Dragon obviously couldn’t apply that to Thunderbird 2. So you have to say “press Control W.” This is where Vocola or another add-on program might come in handy.

Another thing I like about Thunderbird 3 is that there seem to be more buttons included. A few I don’t remember seeing are Reply to List, Archive, and Restart. One of those may be a custom button I forgot about, but I don’t think so. There are also better keyboard shortcuts in Thunderbird’s menus to begin with, so that even if you don’t have a button, you can make a command if you have an add-on program.

Speaking of buttons, there are now automatically included buttons in each e-mail you open. Rather than putting the Delete, Reply, and other like buttons up in the toolbar, Thunderbird has put them above the message, by the subject and recipient lines. (If you use Skip Trash rather than Delete, you will still have to put the Skip button up in the toolbar. You can’t put it within the message.)

One thing I do want to mention is that if you’re using the custom Read button, you want to say “click to select columns to display,” and then uncheck “read.” If you don’t do that, then your Read button won’t work, because Dragon will click the tiny button labeled “click to sort by Read” instead of opening your highlighted message.

Another thing I want to mention is that there is still no total select and say support, but you can still dictate reasonably into a message field — though to be on the safe side, as with anything, you might want to dictate into your word processor or the dictation box first.

I haven’t tried any more add-ons yet, but will report when I do, and when I’ve had more of a chance to use the program. But for now it’s pretty cool.

Features

• Sleek, ultra-slim design fits in a pocket; High-sensitivity GPS receiver for fast satellite lock
• Pre-loaded with City Navigtor NT maps for North America, including more than 6-million name-searchable points of interest
• Speech recognition–speak menu options and keep your eyes on the road

Read Full Detail…

Just the other day Google announced its intentions to automatically generate closed caption files on a select group of YouTube files.  The story quickly made it to the NY Times and all over the blogosphere, as it rightfully should.  The idea is to eventually rollout the capability across YouTube for all users to test.  With 20 hours of video being uploaded to YouTube every minute, that’s a lot of text being created!

At its core, this is a brilliant move by Google to improve YouTube search (and advertising) capabilities.  But Google’s announcement, largely because it’s Google, also puts the accessibility issue in front of the entire country for a change.  Captions are mandated for much of television, but they are only beginning to get some attention on the internet, well until now.  Representative Ed Markey, the same Congressman who made the original push for closed captioning on television, introduced H.R. 3101, the Twenty-first Century Communications and Video Accessibility Act of 2009, during this session of Congress, and it currently has 19 co-sponsors.  This is actually the second attempt at getting a bill passed that would mandate an improved user experience for the hearing impaired.

Thanks to one of the most talked about technology companies of our time, closed captioning is getting attention all over the internet.  Anyone who works with online video is now paying attention to closed captioning.  Not only are we empowering the hearing impaired, but in a virtual world that seems to be driven by search and discovery, video can now be made more “accessible” than ever.

So for a business that is centered on providing high quality, time synchronized transcripts, what does this announcement mean?

Well, it could mean a lot of things.  First, let’s look into this new Google service.  Google will deploy the same technology that powers Google Voice across YouTube to enable the creation of text.  This means they will be using automatic speech recognition (ASR) to create the caption files.  Using ASR on audio and video is not a new concept, but it’s new at this scale.  We’ve commented on our experiences with ASR capbilities in the past.  In fact, we’ve even played with the very engine that will be front and center for the YouTube initiative.

We’ve spoken with many people who have tested ASR solutions.  Usually, if they are talking to us, they weren’t satisfied!  The truth of the matter is that ASR will be good enough for some people, and it won’t be good enough for others.  80% accuracy (at its best and in studio quality recording conditions) leaves a lot to be desired.  In fact, Google even admits that results can be somewhat amusing when they’re off.  On the search front, the most critical keywords tend to be the most unique and, therefore, least common to be recognized accurately.  Google’s announcement does not change that, it just makes an ASR solution easier to use and free to consume.  In many cases, Google has likely provided a medium for people who may never have put captions on their video with the ability to do so with very little effort.  Google has also made the search benefits of captions glaringly obvious.

Ultimately, the organizations that require (or believe in) high quality output for captions and search will be willing to pay for cleaned up text.  There are significant benefits to the high-quality approach, whether it be accurate search results or truly legible transcripts.  Branding is also a critical issue for many organizations who add a text component to their video offering.

We at 3Play Media will continue building high quality solutions that make multimedia more accessible for everyone.  More people than ever are aware of the benefits of captions and time-synchronized transcripts now.  We have some new product launches on the way that will build off these very benefits, and we can’t wait to show the world how their online video experience can be changed forever.

The New York Times today reports that

In the first major step toward making millions of videos on YouTube accessible to deaf and hearing-impaired people, Google unveiled new technologies on Thursday that will automatically bring text captions to many videos on the site.

According to the article, Google’s speech recognition technology, already used in applications such as its voice-to-text phone message service, initially will be applied to largely educational video channels on the YouTube, such as PBS, National Geographic and university stations. Another version of it will be available to regular users, giving them the option of having YouTube caption their videos for them, auto-generating transcripts from the audio.

While aimed at making online videos more accessible to the aurally-challenged, the technology will also be a revelation for the rest of us who want to bask ever more thoroughly in the glow of humanity’s radiant stupidity. Did she just say, “could be a crack head that got hold to the wrong stuff”? Let’s go to the transcript for confirmation. (Yes. I am getting word that, yes, that is what she said.)

Furthermore, imagine the far out search implications. Forget hoping someone’s typed out and time-logged choice quotes from the latest Family Guy (06:24: “I did some poos, I did some poos, I didn’t  mean to.”), or straining to recall in which chapter R. Kelly sings “I thought your name was Mary/ That’s what you said at the party/ Man, this is getting scary/ I’m gonna shoot somebody.” Uploaded videos will be automatically tagged with text files, unleashing a deluge of previously untap-able search reserves. It’s the web’s next logical step, really.

And you thought you spent a lot of time on the computer now.

I’ve mentioned the Custom Toolbar Buttons for using Firefox and Thunderbird with Dragon before, but am plugging it again because there are some navigation buttons I didn’t notice. These buttons would be especially useful for those who don’t use add-on command programs, but they may also simply reduce the syllables of an existing command if that’s important to you.

1. Find Again and Find Previous. This might be helpful for those using typeaheadfind.linksonly rather than Mouseless Browsing. If Firefox focuses on a word with multiple instances, saying “find again/previous” is more fluid than “press Control G” or “press Control Shift G.” However, it will be somewhat slower than the keypress command or a Vocola “find again” command since it uses cursor movement instead of keystrokes.

2. Close Other Tabs. Does exactly what it says, leaving only your current tab.

3. Snap Back. Goes to the first site you visited in the current tab. Useful depending on your browsing habits.