Nortel Networks Corporation

Avaya Inc.

Jason Adelman

Paul Michael Brennan

Lloyd Malcolm Florence

Richard A Weiss

Todd Joseph Wongkee

Lin Lin

Ping Lin

Methods and apparatus for operating on a non-text (voice or fax) message include for each of a number of voice messages, searching each voice message for signal samples having one of a plurality of sets of pre-defined characteristics (such as any seven or ten digit number which is then assumed to be a telephone number). For each signal sample found to have one of the sets of pre-defined characteristics, the found signal sample is associated with an equivalent text string and this text string is stored in an envelope for the message. Based on a user input, the messages may be searched for a given text string and on finding a message envelope containing the given text string, such message may be operated upon.

Methods and apparatus for operating on non-text messages

The present invention provides a means for an end user to customize a speech-based user interface, such as those used in status and control for such applications as messaging, unified communications, automobile control, and many others. The invention consists of two major components: a graphical configuration component, and a run-time component. The graphical configuration component allows a user to quickly and efficiently: specify the information they wish to be included in spoken summaries; restructure command menus or vocabulary structures to meet their personal requirements, moving commonly-used commands to the 'top' of a hierarchy or vocabulary structure, and rarely-used commands to the bottom layers; create additional commands to access information relating to functions external to the system, e.g. access data from the web, or control another device or system, via the web; and save the customization for use by the run-time program. The run time program is embedded into the speech user interface module of the system, and effectively executes the customized speech based user interface. Some existing speech user interface modules may already be 'data driven', thus providing an excellent foundation for implementing the invention.

Speech based status and control user interface customisable by the user

A system and method for speech recognition includes a speaker-independent set of stored word representations derived from speech of many users deemed to be typical speakers and for use by all users, and may further include speaker-dependent sets of stored word representations specific to each user. The speaker-dependent sets may be used to store custom commands, so that a user may replace default commands to customize and simplify use of the system. Utterances from a user which match stored words in either set according to the ordering rules are reported as words.

Speech recognition system and method permitting user customization

Growing community of inventors

Toronto, Canada

Jason Adelman