Ianywhere Solutions, Inc.

Toyota Infotechnology Center Co. Ltd.

Sybase, Inc.

Masahiko Funaki

Babak Hodjat

Benjamin K Reaves

Norikazu Endo

Makoto Kurahashi

Siamak Hodjat

John R Brookes

Kanae Ishii

Mitsuru Shiratori

Junichi Ito

Roughly described, a natural language interpretation system develops its interpretation in response to user input, and also in dependence upon the interpretation, rather than the claims, made in response to prior input. In policy conditions, an agent can check whether any specified agent contributed to the prior interpretation and adjust its claiming of the present input accordingly. In policy actions, a policy or agent can reference the prior interpretation both for conditioning the policy's or agent's contribution to the current interpretation, and also for incorporating part or all of the prior interpretation in the policy's or agent's contribution to the current interpretation. The resulting system thus implements dialoging in a distributed manner as part of the various agents in the interpretation network, and allows much more flexibility to better respond in accordance with users' expectations.

Distributed dialoguing with an agent-oriented software architecture

Shortcut names corresponding to flagged locations or frequently stated destinations, phrases, or sentences are registered, maintained, used, and deregistered in a speech recognition system. If a flagged location or a frequently stated destination, phrase, or sentence is received by the speech recognition system, the user is prompted to provide a shortcut name corresponding to the frequently stated destination, phrase, or sentence in the form of speech, which is then recognized by speech recognition and registered to a personal profile of the user. The shortcut name is also added to a grammar database of the speech recognition system, so that the speech recognition system may recognize the shortcut name itself. The shortcut name may also be deleted, when speech received from a user contains a shortcut name but the user did not intend the shortcut name to mean the corresponding destination, phrase, or sentence.

Shortcut names for use in a speech recognition system

A speech recognition system recognizes an input speech signal by using a first speech recognizer and a second speech recognizer each coupled to a decision module. Each of the first and second speech recognizers outputs first and second recognized speech texts and first and second associated confidence scores, respectively, and the decision module selects either the first or the second speech text depending upon which of the first or second confidence score is higher. The decision module may also adjust the first and second confidence scores to generate first and second adjusted confidence scores, respectively, and select either the first or second speech text depending upon which of the first or second adjusted confidence scores is higher. The first and second confidence scores may be adjusted based upon the location of a speaker, the identity or accent of the speaker, the context of the speech, and the like.

Speech recognition system having multiple speech recognizers

Growing community of inventors

Ichikawa, Japan

Masahiko Funaki