Fonix Introduces VoiceIn 4.0 SE - Speech Recognition SDK

[Kris Kumar]

http://www.fonix.com/page.cfm?name=news&id=1491

"Fonix Corporation, an innovative communications and technology company providing integrated telecommunications services and value-added speech technologies, announces the newest version of our proprietary automatic speech recognition (ASR) SDK. Fonix VoiceIn™ 4.0 SE (Standard Edition) is immediately available to developers looking to implement cutting-edge ASR-driven applications to devices and products."

The Speech Recognition SDK which is compatible with Smartphones and Pocket PCs, will enable the next generation of hands free software to recognize voice commands. Voice recognition for mobile devices makes sense, but the current implementations are mostly limited to dialing phone numbers. Which is often easier to do the regular way. I can't wait for the day when all the mobile apps have not only some kind of speech recognition but can also speak back to me. I want to be able to say "Open Inbox..Read New Mails", while driving my car. 8) For now, I hope the pricing/licensing model is convenient for the ISVs, to quickly adopt this technology. What do you think, are voice recognition and speech capabilities worth the hype? Or do you think it is cumbersome? And what about the social implications, people talking on the phones in a subway is bad enough, and thanks to this technology we will be bothered by people 'talking to' the phone.

[Kris Kumar]

http://www.fonix.com/page.cfm?name=news&id=1491

"Fonix Corporation, an innovative communications and technology company providing integrated telecommunications services and value-added speech technologies, announces the newest version of our proprietary automatic speech recognition (ASR) SDK. Fonix VoiceIn™ 4.0 SE (Standard Edition) is immediately available to developers looking to implement cutting-edge ASR-driven applications to devices and products."

The Speech Recognition SDK which is compatible with Smartphones and Pocket PCs, will enable the next generation of hands free software to recognize voice commands. Voice recognition for mobile devices makes sense, but the current implementations are mostly limited to dialing phone numbers. Which is often easier to do the regular way. I can't wait for the day when all the mobile apps have not only some kind of speech recognition but can also speak back to me. I want to be able to say "Open Inbox..Read New Mails", while driving my car. 8) For now, I hope the pricing/licensing model is convenient for the ISVs, to quickly adopt this technology. What do you think, are voice recognition and speech capabilities worth the hype? Or do you think it is cumbersome? And what about the social implications, people talking on the phones in a subway is bad enough, and thanks to this technology we will be bothered by people 'talking to' the phone.

[Ben]

Originally Posted by Kris Kumar

What do you think, are voice recognition and speech capabilities worth the hype? Or do you think it is cumbersome? And what about the social implications, people talking on the phones in a subway is bad enough, and thanks to this technology we will be bothered by people 'talking to' the phone.

Now I am all about this technology's potential, but I think it is pretty useless right now. I don't care if I can open an application by voice, but I will be thrilled when the day comes that I can compose an email by dictation to my phone! Won't that be the coolest thing ever? I am fine with the screen size of the smartphone, I just hate the data input. Devices with a qwerty thumbpad are generally too big to fit in my pocket comfortably. Yet those without a qwerty thumbpad are generally too slow to write messages, emails, and compose word documents in comfortably. Those are the reasons why we need better speech recognition and relentless improvement and capital thrown at this, so that the accurracy and speed of speech recognition software improves.

It would be nice to have my email, word documents, or eBooks read to me while driving, as well. I look forward to that, too, but it has to be a pleasant female voice to listen to . . .

[Kris Kumar]

I agree Voice Dictation will overcome the keypad input limitation on the Smartphones.

As for the female voice...well I have GPS navigation system in my car, and its female voice makes my girl friend mad :-)

[Mike Temporale]

Originally Posted by Kris Kumar

As for the female voice...well I have GPS navigation system in my car, and its female voice makes my girl friend mad :-)

I imagine that this will be "skin-able" one day. Allowing you to pick the voice that you like best.

[John Cody]

Originally Posted by Mike Temporale

I imagine that this will be "skin-able" one day. Allowing you to pick the voice that you like best.

Your wish is granted...

http://www.elanspeech.com/randd/saysotechno.html

(see "Elan Sayso™ can re-create anyone's voice" and "How are new voices created?)

[Mike Temporale]

Originally Posted by John Cody

Your wish is granted...

http://www.elanspeech.com/randd/saysotechno.html

(see "Elan Sayso™ can re-create anyone's voice" and "How are new voices created?)

Cool!

The first step in creating a new voice for Sayso technology is to record a speaker reading a selected corpus of text collated from some 600,000 printed pages. The corpus is selected for its rich phonetic representation and wide-ranging prosodic situations.

Ouch! that's not as easy as I imagined it would be. But it's an understandable process.

[Kris Kumar]

600,000 Printed Pages 8O

Assuming 300 page per book, I can finish 2 books in a month (bedtime reading).

600 pages a month, 12 months in a year = 7200 pages in a year.

It will take me 83 8O years to complete 600,000 pages!