Subband-based audio calibration
A technique for subband-based audio calibration includes receiving a first audio signal captured at a microphone and a second audio signal provided to a speaker. The technique continues by synchronizing the second audio signal with the first audio signal to generate a synchronized audio signal and transforming the synchronized audio signal to a first subband representation and the first audio signal to a second subband representation. The technique then includes calculating a transfer function associated with the speaker and the microphone based on the first subband representation and the second subband representation.
1. A method, comprising:
receiving a first audio signal captured at a microphone and a second audio signal provided to a speaker;
synchronizing the second audio signal with the first audio signal to generate a synchronized audio signal;
transforming the synchronized audio signal to a first subband representation and the first audio signal to a second subband representation; and
calculating a transfer function associated with the speaker and the microphone based on the first subband representation, the second subband representation, wherein a rate of updating the transfer function is based on a coherence between at least a portion of the first subband representation and at least a portion of the second subband representation.
2. The method of claim 1 , wherein synchronizing the second audio signal comprises:
determining a delay between the first audio signal and the second audio signal; and
processing the second audio signal based on the delay to generate the synchronized audio signal.
3. The method of claim 1 , wherein the first subband representation comprises a first set of spectral bands and the second subband representation comprises a second set of spectral bands that correspond respectively to the first set of spectral bands, and further comprising calculating a coherence for each pair of corresponding spectral bands of the first set of spectral bands and the second set of spectral bands.
4. The method of claim 3 , further comprising updating, based at least in part on the coherence, a first partial transfer function associated with a first spectral band of the first subband representation and the second subband representation.
5. The method of claim 4 , further comprising:
measuring acoustic background noise, wherein updating the partial transfer functions is performed at a rate that is based, at least in part, on a comparison between one or more bands of the first subband representation and the acoustic background noise.
6. The method of claim 4 , wherein updating is further based on a smoothing factor that depends, at least in part, on the coherence.
7. The method of claim 6 , wherein a functional relation between the smoothing factor and the coherence is based, at least in part, on psycho-acoustic parameters.
8. The method of claim 1 , wherein the second audio signal provided to the speaker is filtered based at least in part on the transfer function.
9. The method of claim 1 , further comprising:
deriving an equalization filter based at least in part on the transfer function; and
applying the equalization filter to the second audio signal.
10. The method of claim 1 , wherein the first subband representation comprises a first set of spectral bands and the second subband representation comprises a second set of spectral bands that correspond respectively to the first set of spectral bands, and wherein calculating the transfer function further comprises calculating a partial transfer function for each pair of corresponding spectral bands of the first and second set of spectral bands.
11. The method of claim 1 , wherein calculating the transfer function comprises:
determining a signal to noise ratio for an audio signal with respect to audio noise;
estimating a transfer function between a speaker and a microphone based, at least in part, on two previous transfer functions between the speaker and the microphone; and
in an iterative process, updating an estimate of the transfer function between the speaker and the microphone based, at least in part, on another two previous transfer functions between the speaker and the microphone, wherein a rate of the iterative process of updating the estimate of the transfer function depends, at least in part, on the signal to noise ratio.
12. The method of claim 11 , wherein the rate of the iterative process increases when the signal to noise ratio increases and the rate of the iterative process decreases when the signal to noise ratio decreases.
13. The method of claim 11 , further comprising calculating a coherence between each spectral band of the first and the second subband representations, wherein the rate of the iterative process of updating the estimate of the transfer function further depends, at least in part, on the coherence.
14. The method of claim 1 , wherein the microphone and the speaker are included in a system configured for augmented reality or virtual reality, wherein the system further includes a display source.
15. A non-transitory computer readable medium storing instructions that, when executed by a processor, cause the processor to perform the steps of:
receiving a first audio signal captured at a microphone and a second audio signal provided to a speaker;
synchronizing the second audio signal with the first audio signal to generate a synchronized audio signal;
transforming the synchronized audio signal to a first subband representation and the first audio signal to a second subband representation; and
calculating a transfer function associated with the speaker and the microphone based on the first subband representation, the second subband representation, wherein a rate of updating the transfer function is based on a coherence between at least a portion of the first subband representation and at least a portion of the second subband representation.
16. The non-transitory computer readable medium of claim 15 , wherein the first subband representation comprises a first set of spectral bands and the second subband representation comprises a second set of spectral bands that correspond respectively to the first set of spectral bands, and further comprising calculating a coherence for each pair of corresponding spectral bands of the first set of spectral bands and the second set of spectral bands.
17. The non-transitory computer readable medium of claim 16 , further comprising updating, based at least in part on the coherence, a first partial transfer function associated with a first spectral band of the first subband representation and the second subband representation.
18. The non-transitory computer readable medium of claim 16 , wherein calculating the transfer function comprises:
determining a signal to noise ratio for an audio signal with respect to audio noise;
estimating a transfer function between a speaker and a microphone based, at least in part, on two previous transfer functions between the speaker and the microphone; and
in an iterative process, updating an estimate of the transfer function between the speaker and the microphone based, at least in part, on another two previous transfer functions between the speaker and the microphone, wherein a rate of the iterative process of updating the estimate of the transfer function depends, at least in part, on the signal to noise ratio.
19. The non-transitory computer readable medium of claim 18 , wherein the rate of the iterative process increases when the signal to noise ratio increases and the rate of the iterative process decreases when the signal to noise ratio decreases.
20. A system, comprising:
a calibration module that:
receives a first audio signal captured at a microphone and a second audio signal provided to a speaker,
synchronizes the second audio signal with the first audio signal to generate a synchronized audio signal,
transforms the synchronized audio signal to a first subband representation and the first audio signal to a second subband representation, and
calculates a transfer function associated with the speaker and the microphone based on the first subband representation, the second subband representation, wherein a rate of updating the transfer function is based on a coherence between at least a portion of the first subband representation and at least a portion of the second subband representation.
21. The system of claim 20 , wherein the calibration module calculates the transfer function by:
determining a signal to noise ratio for an audio signal with respect to audio noise;
estimating a transfer function between a speaker and a microphone based, at least in part, on two previous transfer functions between the speaker and the microphone; and
in an iterative process, updating an estimate of the transfer function between the speaker and the microphone based, at least in part, on another two previous transfer functions between the speaker and the microphone, wherein a rate of the iterative process of updating the estimate of the transfer function depends, at least in part, on the signal to noise ratio,
wherein the rate of the iterative process increases when the signal to noise ratio increases and the rate of the iterative process decreases when the signal to noise ratio decreases.
22. The method of claim 1 , wherein the rate of updating the transfer function is inversely proportional to a magnitude of the coherence.
23. The method of claim 1 , wherein the transfer function and the coherence are unique to a first spectral band of a plurality of spectral bands of the first subband representation and the second subband representation.