MATLAB script for auxiliary-function-based independent vector analysis with iterative source steering (AuxIVA-ISS) and its application to blind audio source separation.
- input [dir]: includes test audio signals (reverberation time is around 300 ms)
- AuxIVAISS.m: function of AuxIVA-ISS with pre- and post-processing (STFT, whitening, back projection, and ISTFT)
- backProjection.m: back projection technique (fixing frequency-wise scales)
- ISTFT.m: inverse short-time Fourier transform
- main.m: main script with parameter settings
- STFT.m: short-time Fourier transform
- whitening.m: applying principal component analysis for decorrelating observed multichannel signal
Coded by Daichi Kitamura
- T. Kim, H. T. Attias, S.-Y. Lee, and T.-W. Lee, "Blind source separation exploiting higher-order frequency dependencies," IEEE Trans. Audio, Speech, and Language Processing, vol. 15, no. 1, pp. 70–79, 2007.
- N. Ono, "Stable and fast update rules for independent vector analysis based on auxiliary function technique", IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), pp.189-192, 2011.
- S. Robin and N. Ono, "Fast and stable blind source separation with rank-1 updates", IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp.236-240, 2020.
You can find Python script of AuxIVA-IP and AuxIVA-ISS in Pyroomacoustics: https://pyroomacoustics.readthedocs.io/en/pypi-release/pyroomacoustics.bss.auxiva.html#module-pyroomacoustics.bss.auxiva