Reverberation removal
View Patent ↗A method of removing reverberation from audio signals is disclosed. The method comprises spectro-temporally analyzing the first audio signal and the second audio signal to derive an energy function of time for a plurality of frequency bands. The method further comprises determining a delay stability between the energy function of time for the first audio signal and the second audio signal in each band, determining a gain function in each band based on the delay stability, adjusting the energy of the first audio signal and the second audio signal using the gain function within each band, and resynthesizing audio signals from the energy in each band of the first audio signal and the second audio signal.
1. A method of removing reverberation from an audio signal comprising:
spectro-temporally analyzing the audio signal to derive an energy function of time for a plurality of frequency bands,
filtering for each frequency band the energy function of time with a first low pass filter,
filtering for each frequency band the energy function of time with a second low pass filter where the second low pass filter has a longer time constant than the first low pass filter,
subtracting for each frequency band the output of the second low pass filter from the output of the first low pass filter,
resynthesizing an audio signal from the output signal of the subtraction in each frequency band, and
mixing the resynthesized audio signal with the original signal.
2. A method of removing reverberation from a signal as in claim 1 wherein the audio signal is digital.
3. A method of removing reverberation from a signal as in claim 1 wherein the audio signal is analog.
4. A method of removing reverberation from a signal as in claim 1 wherein spectro-temporally analyzing the audio signal uses a Fourier Transform.
5. A method of removing reverberation from a signal as in claim 1 wherein spectro-temporally analyzing the audio signal uses a Fast Fourier Transform.
6. A method of removing reverberation from a signal as in claim 1 wherein spectro-temporally analyzing the audio signal uses Gammatone filter banks.
7. A method of removing reverberation from a signal as in claim 1 wherein spectro-temporally analyzing the audio signal uses a wavelet Transform.
8. A method of removing reverberation from a signal as in claim 1 wherein spectro-temporally analyzing the audio signal uses a cochlear model.
9. A method of removing reverberation from an audio signal comprising:
spectro-temporally analyzing the audio signal to derive an energy function of time for a plurality of frequency bands,
filtering for each frequency band the energy function of time with a first low pass filter,
filtering for each frequency band the energy function of time with a second low pass filter where the second low pass filter has a longer time constant than the first low pass filter,
subtracting for each frequency band the output of the second low pass filter from the output of the first low pass filter,
mixing the output signal of the subtraction with the original energy function of time for each frequency band, and
resynthesizing an audio signal from the energy in each frequency band.
10. A method of removing reverberation from a signal as in claim 9 wherein the audio signal is digital.
11. A method of removing reverberation from a signal as in claim 9 wherein the audio signal is analog.
12. A method of removing reverberation from a signal as in claim 9 wherein spectro-temporally analyzing the audio signal uses a Fourier Transform.
13. A method of removing reverberation from a signal as in claim 9 wherein spectro-temporally analyzing the audio signal uses a Fast Fourier Transform.
14. A method of removing reverberation from a signal as in claim 9 wherein spectro-temporally analyzing the audio signal uses Gammatone filter banks.
15. A method of removing reverberation from a signal as in claim 9 wherein spectro-temporally analyzing the audio signal uses a wavelet Transform.
16. A method of removing reverberation from a signal as in claim 9 wherein spectro-temporally analyzing the audio signal uses a cochlear model.