Recognizing speech in the presence of additional audio
View Patent ↗The technology described in this document can be embodied in a computer-implemented method that includes receiving, at a processing system, a first signal including an output of a speaker device and an additional audio signal. The method also includes determining, by the processing system, based at least in part on a model trained to identify the output of the speaker device, that the additional audio signal corresponds to an utterance of a user. The method further includes initiating a reduction in an audio output level of the speaker device based on determining that the additional audio signal corresponds to the utterance of the user.
1. A computer implemented method comprising:
providing a text-to-speech prompt for output;
receiving, at a processing system, particular audio data encoding (i) at least a portion of the text-to-speech prompt and (ii) a user utterance;
providing the particular audio data to an additional audio activity detector comprising a model that is trained, using training audio data comprising text-to-speech prompts, to identify whether given audio data comprises additional audio other than a text-to-speech prompt;
receiving, from the additional audio activity detector, data indicating that the particular audio data comprises additional audio other than the text-to-speech prompt; and
in response to receiving the data indicating that the particular audio data comprises additional audio other than the text-to-speech prompt, initiating a reduction in an audio output level of the text-to-speech prompt.
2. The method of claim 1 , wherein initiating a reduction in an audio output level of the text-to-speech prompt comprises interrupting output of the text-to-speech prompt.
3. The method of claim 1 , further comprising:
generating a transcription corresponding to the user utterance.
4. The method of claim 1 , further comprising:
obtaining a first vector corresponding to the text-to-speech prompt;
comparing the first vector to a second vector corresponding to the model; and
determining that the particular audio data comprises additional audio other than the text-to-speech prompt based on a result of the comparison satisfying a threshold.
5. The method of claim 1 , further comprising:
obtaining a first vector corresponding to the text-to-speech prompt; and
determining that the particular audio data comprises additional audio other than the text-to-speech prompt based on the first vector satisfying a threshold.
6. The method of claim 1 , wherein the model is an i-vector based model.
7. The method of claim 1 , wherein the model is a neural network based model.
8. The method of claim 1 , wherein the model jointly represents a user voice and the text-to-speech prompt.
9. A system comprising:
one or more computers and one or more storage devices storing instructions that are operable, when executed by the one or more computers, to cause the one or more computers to perform operations comprising:
providing a text-to-speech prompt for output;
receiving, at a processing system, particular audio data encoding (i) at least a portion of the text-to-speech prompt and (ii) a user utterance;
providing the particular audio data to an additional audio activity detector comprising a model that is trained, using training audio data comprising text-to-speech prompts, to identify whether given audio data comprises additional audio other than a text-to-speech prompt;
receiving, from the additional audio activity detector, data indicating that the particular audio data comprises additional audio other than the text-to-speech prompt; and
in response to receiving the data indicating that the particular audio data comprises additional audio other than the text-to-speech prompt, initiating a reduction in an audio output level of the text-to-speech prompt.
10. The system of claim 9 , wherein initiating a reduction in an audio output level of the text-to-speech prompt comprises interrupting output of the text-to-speech prompt.
11. The system of claim 9 , further comprising:
generating a transcription corresponding to the user utterance.
12. The system of claim 9 , further comprising:
obtaining a first vector corresponding to the text-to-speech prompt;
comparing the first vector to a second vector corresponding to the model; and
determining that the particular audio data comprises additional audio other than the text-to-speech prompt based on a result of the comparison satisfying a threshold.
13. The system of claim 9 , further comprising:
obtaining a first vector corresponding to the text-to-speech prompt; and
determining that the particular audio data comprises additional audio other than the text-to-speech prompt based on the first vector satisfying a threshold.
14. The system of claim 9 , wherein the model is an i-vector based model.
15. The system of claim 9 , wherein the model is a neural network based model.
16. The system of claim 9 , wherein the model jointly represents a user voice and the text-to-speech prompt.
17. A non-transitory computer readable storage device storing instructions executable by one or more processors which, upon such execution, cause the one or more processors to perform operations comprising:
providing a text-to-speech prompt for output;
receiving, at a processing system, particular audio data encoding (i) at least a portion of the text-to-speech prompt and (ii) a user utterance;
providing the particular audio data to an additional audio activity detector comprising a model that is trained, using training audio data comprising text-to-speech prompts, to identify whether given audio data comprises additional audio other than a text-to-speech prompt;
receiving, from the additional audio activity detector, data indicating that the particular audio data comprises additional audio other than the text-to-speech prompt; and
in response to receiving the data indicating that the particular audio data comprises additional audio other than the text-to-speech prompt, initiating a reduction in an audio output level of the text-to-speech prompt.
18. The computer readable storage device of claim 17 , wherein initiating a reduction in an audio output level of the text-to-speech prompt comprises interrupting output of the text-to-speech prompt.
19. The computer readable storage device of claim 17 , further comprising:
generating a transcription corresponding to the user utterance.
20. The computer readable storage device of claim 17 , further comprising:
obtaining a first vector corresponding to the text-to-speech prompt;
comparing the first vector to a second vector corresponding to the model; and
determining that the particular audio data comprises additional audio other than the text-to-speech prompt based on a result of the comparison satisfying a threshold.