Browse Prior Art Database

Two-Way Voice Recognition Instant Messenger (VRIM) with Network Voice Conversion Flexibility

IP.com Disclosure Number: IPCOM000023999D
Publication Date: 2004-Apr-01
Document File: 9 page(s) / 108K

Publishing Venue

The IP.com Prior Art Database

Abstract

To develop a 2-way voice-to-text and text-to-voice network conversion system. Communications can be sent via telephone. VoIP or text, and converted to the preferred format of the recipient based on specified delivery preferences instantly. This will involve platform development and possible voice-recognition software purchases and alliances. The key differentiator is the instant messaging conversion.

This text was extracted from a Microsoft Word document.
This is the abbreviated version, containing approximately 10% of the total text.

         Sony Corporation

         Sony Electronics Inc.

         IPD Case #50R4751

Title:

Two-Way Voice Recognition Instant Messenger (VRIM) with Network Voice Conversion Flexibility        

Abstract:

To develop a 2-way voice-to-text and text-to-voice network conversion system. Communications can be sent via telephone. VoIP or text, and converted to the preferred format of the recipient based on specified delivery preferences instantly. This will involve platform development and possible voice-recognition software purchases and alliances. The key differentiator is the instant messaging conversion.

Inventors:

Patrick F. Fitzpatrick

Natasha King

Background Information:

Useful technologies which need software/API to incorporate as instant message system: There are currently no specific tools enabling conversion of message from text to voice and voice to text based on specified preferences through a network which is device agnostic. This future message format would be an automatic conversion based on pre-specified preferences - and would actively and instantly deliver to the user based upon their preference for voice or text. There are, however, many voice recognition technologies which can be useful to incorporate including:

PC Dictation - Packaged Software

·         Dragon Systems makes a dictation engine that can be adapted for specific vocabularies.

·         IBM Via Voice - web site for PC-based dictation application.

·         Lernout & Hauspie sells a line of PC-based dictation software.

PC Recognition Engines

·         The AT&T Advanced Speech Products Group offers WATSON. with SAPI-compatible speech recognition and speech synthesis, as well as speaker verification technologies.

·         Verbex Voice Systems was one of the first providers of continuous speech recognition technology.

Their applications include work with the UPS, L'Eggs. Canada Post, and Medical Labs. They also have OEM software on the PC platform, Listen for Windows.

·         Microsoft Research Speech Technology Group is developing the Whisper engine.

·         Command Corporation makes the IN CUBE recognizer.

·         SRI Corp's STAR Lab has 25 people developing a wideband, continuous speech recognizer called DECIPHER and has been doing some interesting work with a new engine called Corona (see here for an in-house experiment with using Corona for command and control of Unix).

·         Apple's Plain Talk incorporates speech recognition and synthesis into the Mac as.

·         The DDLinux page covers speech engines that have been ported to Linux and even lists a handful of open source engines.

Telephony and Call Center Engines

·         Here are the demo pages from several major vendors: Speech Works, VCS, Nuance, Locus, and Vocalis.

·         Northern Telecom is developing speech recognition engines.

·         Speech Works (previously AL Tech) develops speech recognition engine technology for over-the­telephone customer service and transactions.

·         Voice Control Systems has acquired several other speech recognition companies including PureSpeech. They offer a broad line of telephony-based...