Enhanced accuracy for speech recognition grammars
AssigneeAT&T Intellectual Property II, L.P.
InventorsHarry Blanchard, Steven Lewis, Shankarnarayan Sivaprasad, Lan Zhang
Life of this patent
Estimating the term…
Abstract
Disclosed herein are methods and systems for recognizing speech. A method embodiment comprises comparing received speech with a precompiled grammar based on a database and if the received speech matches data in the precompiled grammar then returning a result based on the matched data. If the received speech does not match data in the precompiled grammar, then dynamically compiling a new grammar based only on new data added to the database after the compiling of the precompiled grammar. The database may comprise a directory of names.
Claims20
A method comprising:
- comparing, via a processor, received speech with a precompiled grammar to yield a match result, wherein the precompiled grammar is based on a database;
- and
- when the match result is below a threshold, comparing the received speech with a dynamically compiled grammar that is based only on new data added to the database after the precompiled grammar is compiled.
A system comprising:
- a processor;
- and
- a computer-readable medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
- comparing received speech with a precompiled grammar to yield a match result, wherein the precompiled grammar is based on a database;
- and
- when the match result is below a threshold, comparing the received speech with a dynamically compiled grammar that is based only on new data added to the database after the precompiled grammar is compiled.
A computer-readable medium having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
- comparing received speech with a precompiled grammar to yield a match result, wherein the precompiled grammar is based on a database;
- and
- when the match result is below a threshold, comparing the received speech with a dynamically compiled grammar that is based only on new data added to the database after the precompiled grammar is compiled.
Dependencies are inferred from the claim text.