Speech recognition with parallel recognition tasks

A computer system comprising:

• one or more computing devices;
• an interface of the one or more computing devices that is programmed to receive an audio signal;
• and
• a plurality of speech recognition systems that initiate a plurality of speech recognition tasks for the audio signal;
• a recognition managing module that is programmed to:
• detect that a portion of the plurality of speech recognition systems have completed their respective speech recognition tasks which comprise a completed portion of the plurality of speech recognition tasks, wherein a remaining portion of the plurality of speech recognition systems have not completed their respective speech recognition tasks and are still processing their respective speech recognition tasks,
• obtain recognition results and confidence values for the completed portion of the plurality of speech recognition tasks, wherein the recognition results identify one or more candidate representations of the audio signal and the confidence values identify one or more probabilities that the recognition results are correct,
• generate one or more combined confidence values for the recognition results based on the recognition results and the confidence values for the completed portion of the plurality of speech recognition tasks,
• determine whether at least one of the one or more combined confidence values is greater than or equal to a threshold confidence value, and
• in response to determining that the at least one of the one or more combined confidence values is greater than or equal to the threshold confidence value and before the remaining portion of the plurality of speech recognition systems have completed their respective speech recognition tasks, provide through the interface a final recognition result for the audio signal based on the recognition results and the one or more combined confidence values.

A computer program product embodied in a computer readable storage device storing instructions that, when executed, cause one or more computing devices to perform operations comprising:

• receiving an audio signal;
• initiating a plurality of speech recognition tasks for the audio signal, the plurality of speech recognition tasks running on a plurality of speech recognitions systems;
• detecting that a portion of the plurality of the speech recognition systems have completed their respective speech recognition tasks and have yielded a completed portion of the plurality of speech recognition tasks, wherein a remaining portion of the plurality of the speech recognition systems have not completed their respective speech recognition tasks and are still processing their respective speech recognition tasks;
• obtaining recognition results and confidence values for the completed portion of the plurality of speech recognition tasks, wherein the recognition results identify one or more candidate representations of the audio signal and the confidence values identify one or more probabilities that the recognition results are correct;
• generating one or more combined confidence values for the recognition results based on the recognition results and the confidence values for the completed portion of the plurality of speech recognition tasks;
• determining whether at least one of the one or more combined confidence values is greater than or equal to a threshold confidence value;
• and
• in response to determining that the at least one of the one or more combined confidence values is greater than or equal to the threshold confidence value and before the remaining portion of the plurality of speech recognition systems have completed their respective speech recognition tasks, providing a final recognition result for the audio signal based on the recognition results and the one or more combined confidence values.