Stored phrase reutilization when testing speech recognition

摘要

A set of audio phrases and corresponding phrase characteristics can be maintained, such as in a database. The phrase characteristics can include a translation of speech in the associated audio phrase. A finite state grammar that includes a set of textual phrases can be received. A software algorithm can execute to compare the set of textual phrases against the translations associated with the maintained audio phrases. A result of the software algorithm execution can be produced, where the result indicates phrase coverage for the finite state grammar based upon the audio phrases.

主权项

1. A method of evaluating a plurality of stored audio phrases and at least one finite state grammar, wherein the at least one finite state grammar defines at least a first plurality of text phrases and wherein the plurality of stored audio phrases correspond to a second plurality of text phrases, the method comprising:
determining, for each one text phrase of the first plurality of text phrases and based at least in part on phrase characteristics associated with the plurality of stored audio phrases, whether there is an audio phrase of the plurality of stored audio phrases that corresponds to the one text phrase, wherein the determining is performed using at least one programmed processor; and identifying a phrase coverage for the at least one finite state grammar based at least in part on the determining, the phrase coverage indicating whether there are one or more text phrases of the first plurality of text phrases defined by the at least one finite state grammar to which none of the plurality of stored audio phrases corresponds.