Systems and methods for recognizing a speech of a speaker
Systems, methods, and computer readable media comprising instructions executable by a processor, for recognizing speech within a received audio signal segment the audio signal to isolate the speech based on a speaker audio profile, determine from the audio signal a command, a first score reflecting confidence in determining the command, and a second score reflecting a potential error in determining the command, and cause the command to be executed if the first score is above a first threshold value and the second score is below a second threshold value.
1. A computer-implemented method applied to a computing device for recognizing a command within an audio signal, the method comprising:
identifying, by a processor, a speaker from the audio signal;
generating a speech spectrogram of the audio signal;
separating, using a computer-based machine learning model, the audio signal to obtain a segment of the audio signal containing a speech associated with the identified speaker based on the speech spectrogram of the audio signal and an audio profile associated with the identified speaker;
determining, from the segment of the audio signal, the command from the speech associated with the identified speaker;
determining, from the segment of the audio signal, a score reflecting a value of confidence that the command determined is a valid command; and
in response to determining that the score is above a first threshold value, causing the command to be executed at the computing device.
2. The computer-implemented method of claim 1 , further comprising:
prior to separating the audio signal, reducing background noise from the audio signal based on the audio profile associated with the identified speaker.
3. The computer-implemented method of claim 1 , wherein the separating the audio signal comprises separating the audio signal based on at least one of an audio waveform, a cadence of the speech, a pitch of the speech, loudness of the speech, or vocabulary of the speech.
4. The computer-implemented method of claim 1 , further comprising:
receiving video data of the speaker producing the audio signal; and
wherein the separating the audio signal comprises separating the audio signal based on a correlation of the audio signal and the video data.
5. The computer-implemented method of claim 1 , further comprising:
receiving vibration data of the speaker producing the audio signal; and
wherein the separating the audio signal comprises separating the audio signal based on a correlation of the audio signal and the vibration data.
6. The computer-implemented method of claim 1 , further comprising in response to determining that the score is at or below a first threshold value:
requesting confirmation speech to validate the command from the speech associated with the identified speaker, wherein the requesting comprises, prompting the speaker for a second audio signal validating the command;
receiving, the second audio signal containing the confirmation speech;
determining whether the confirmation speech validates the command from the speech associated with the identified speaker; and
upon determining that the confirmation speech validates the command, causing the command to be executed.
7. The computer-implemented method of claim 6 , wherein the prompting the speaker comprises prompting the speaker using synthesized speech.
8. A non-transitory, computer-readable medium storing a set of instructions that, when executed by a processor in a computing device, cause:
identifying a speaker from an audio signal;
generating a speech spectrogram of the audio signal;
separating, using a computer-based machine learning model, the audio signal to obtain a segment of the audio signal containing a speech associated with the identified speaker based on the speech spectrogram of the audio signal and an audio profile associated the identified speaker;
determining, from the segment of the audio signal, a command from the speech associated with the identified speaker;
determining, from the segment of the audio signal, a score reflecting a value of confidence that the command determined is a valid command; and
in response to determining that the score is above a first threshold value, causing the command to be executed at the computing device.
9. The non-transitory, computer-readable medium of claim 8 , storing further instructions that, when executed by the processor, cause prior to separating the audio signal, reducing background noise from the audio signal based on the audio profile associated with the identified speaker.
10. The non-transitory, computer-readable medium of claim 8 , wherein the separating the audio signal comprises separating the audio signal based at least on one of an audio waveform, a cadence of the speech, a pitch of the speech, loudness of the speech, or vocabulary of the speech.
11. The non-transitory, computer-readable medium of claim 8 , storing further instructions that, when executed by the processor, cause:
receiving video data of the speaker producing the audio signal; and
wherein the separating the audio signal comprises separating the audio signal based on a correlation of the audio signal and the video data.
12. The non-transitory, computer-readable medium of claim 8 , storing further instructions that, when executed by the processor, cause:
receiving vibration data of the speaker producing the audio signal; and
wherein the separating the audio signal comprises separating the audio signal based on a correlation of the audio signal and the vibration data.
13. The non-transitory, computer-readable medium of claim 8 , storing further instructions that, when executed by the processor, cause, in response to determining that the score is at or below a first threshold value:
requesting confirmation speech to validate the command from the speech associated with the identified speaker, wherein the requesting comprises, prompting the speaker for a second audio signal validating the command;
receiving, the second audio signal containing the confirmation speech;
determining whether the confirmation speech validates the command from the speech associated with the identified speaker; and
upon determining that the confirmation speech validates the command, causing the command to be executed.
14. The non-transitory, computer-readable medium of claim 13 , wherein the prompting the speaker comprises prompting the speaker using synthesized speech.
15. A system in a computing device for recognizing speech in a computing device, comprising:
a processor; and
a memory operatively connected to the processor and storing a set of instructions that, when executed by the processor, cause:
identifying a speaker from an audio signal;
generating a speech spectrogram of the audio signal;
separating, using a computer-based machine learning model, the audio signal to obtain a segment of the audio signal containing a speech associated with the identified speaker based on the speech spectrogram of the audio signal and an audio profile associated with the identified speaker;
determining, from the segment of the audio signal, a command from the speech associated with the identified speaker;
determining, from the segment of the audio signal, a score reflecting a value of confidence that the command determined is a valid command; and
in response to determining that the score is above a first threshold value, causing the command to be executed at the computing device.
16. The system of claim 15 , storing further instructions that, when executed by the processor, cause prior to separating the audio signal, reducing background noise from the audio signal based on the audio profile associated with the identified speaker.
17. The system of claim 15 , wherein the separating the audio signal comprises separating the audio signal based at least on one of an audio waveform, a cadence of the speech, a pitch of the speech, loudness of the speech, or vocabulary of the speech.
18. The system of claim 15 , wherein the memory stores further instructions that, when executed by the processor, cause:
receiving video data of the speaker producing the audio signal; and
wherein the separating the audio signal comprises separating the audio signal based on a correlation of the audio signal and the video data.
19. The system of claim 15 , wherein the memory stores further instructions that, when executed by the processor, cause:
receiving vibration data of the speaker producing the audio signal; and
wherein the separating the audio signal comprises separating the audio signal based on a correlation of the audio signal and the vibration data.
20. The system of claim 15 , storing further instructions that, when executed by the processor, cause, in response to determining that the score is at or below a first threshold value:
requesting confirmation speech to validate the command from the speech associated with the identified speaker, wherein the requesting comprises, prompting the speaker for a second audio signal validating the command;
receiving, the second audio signal containing the confirmation speech;
determining whether the confirmation speech validates the command from the speech associated with the identified speaker; and
upon determining that the confirmation speech validates the command, causing the command to be executed.