System and method for processing sound signals implementing a spectral motion transform
A system and method are provided for processing sound signals. The processing may include identifying individual harmonic sounds represented in sound signals, determining sound parameters of harmonic sounds, classifying harmonic sounds according to source, and/or other processing. The processing may include transforming the sound signals (or portions thereof) into a space which expresses a transform coefficient as a function of frequency and chirp rate. This may facilitate leveraging of the fact that the individual harmonics of a single harmonic sound may have a common pitch velocity (which is related to the chirp rate) across all of its harmonics in order to distinguish an the harmonic sound from other sounds (harmonic and/or non-harmonic) and/or noise.
1. A system configured to process a sound signal, the system comprising:
one or more physical computer processors configured by machine-readable instructions to:
transform a signal portion of a sound signal into a frequency-chirp domain, wherein the signal portion corresponds to a time window that is larger than a sampling frequency of the signal potion, wherein the frequency-chirp domain representation of the signal portion specifies a transform coefficient as a function of frequency and fractional chirp rate for the signal portion, and wherein the fractional chirp rate is chirp rate divided by frequency; and
generate, based on the transformed signal portion, a first sound parameter of an individual harmonic sound represented in the sound signal.
2. The system of claim 1 , wherein the one or more physical computer processors are further configured to define the signal portion within the sound signal.
3. The system of claim 1 , wherein the one or more physical computer processors are further configured to generate, based on the transformed signal portion, the first sound parameter of a second harmonic sound represented in the sound signal, the second harmonic sound being different than the individual harmonic sound.
4. The system of claim 1 , wherein the one or more physical computer processors are further configured to generate, based on the transformed signal portion, a second sound parameter of the individual harmonic sound represented in the sound signal, the second sound parameter being a different parameter than the first sound parameter.
5. The system of claim 1 , wherein the first sound parameter includes one or both of pitch and/or chirp rate.
6. The system of claim 1 , wherein the transform of the signal portion is obtained by applying a set of filters to the signal portion, wherein individual filters in the set of filters correspond to different frequency and chirp rate pairs.
7. The system of claim 1 , wherein the transform coefficient comprises a real component and an imaginary component.
8. The system of claim 1 , wherein the transform coefficient is the modulus squared of a real component and an imaginary component.
9. The system of claim 1 , wherein the one or more physical computer processors are further configured to identify sounds generated by separate sources based on the transform coefficient.
10. A method of processing a sound signal, the method comprising:
transforming a signal portion of a sound signal into a frequency-chirp domain, wherein the signal portion corresponds to a time window that is larger than a sampling frequency of the signal potion, wherein the frequency-chirp domain representation of a given signal portion specifies a transform coefficient as a function of frequency and fractional chirp rate for the signal portion, and wherein fractional chirp rate is chirp rate divided by frequency; and
generating, based on the transformed signal portion, a first sound parameter of an individual harmonic sound represented in the sound signal.
11. The method of claim 10 , further comprising defining the signal portion within the sound signal.
12. The method of claim 10 , further comprising generating, based on the transformed signal portion, the first sound parameter of a second harmonic sound represented in the sound signal, the second harmonic sound being different than the individual harmonic sound.
13. The method of claim 10 , further comprising determining, based on the transformed signal portions, a second sound parameter of the individual harmonic sound represented in the sound signal, the second sound parameter being a different parameter than the first sound parameter.
14. The method of claim 10 , wherein the first sound parameter includes one or both of pitch and/or chirp rate.
15. The method of claim 10 , wherein transforming the individual signal portion comprises applying a set of filters to the individual signal portion, wherein individual filters in the set of filters correspond to different frequency and chirp rate pairs.
16. The method of claim 10 , wherein the transform coefficient comprises a real component and an imaginary component.
17. The method of claim 10 , wherein the transform coefficient is the modulus squared of a real component and an imaginary component.
18. The method of claim 10 , further comprising identifying sounds generated by separate sources based on the transform coefficients.
19. Non-transitory, machine-readable electronic storage media that stores processor-executable instructions for performing a method of processing a sound signal, the method comprising:
transforming a signal portion of a sound signal into a frequency-chirp domain, wherein the signal portion corresponds to a time window that is larger than a sampling frequency of the signal potion, wherein the frequency-chirp domain representation of a given signal portion specifies a transform coefficient as a function of frequency and fractional chirp rate for the signal portion, and wherein fractional chirp rate is chirp rate divided by frequency; and
generating, based on the transformed signal portion, a first sound parameter of an individual harmonic sound represented in the sound signal.