Title :
Informed Source Separation from compressed mixtures using spatial wiener filter and quantization noise estimation
Author :
Shuhua Zhang ; Girin, Laurent ; Liutkus, Antoine
Author_Institution :
GIPSA-Lab., Grenoble-INP, Grenoble, France
Abstract :
In a previous work, we proposed an Informed Source Separation system based on Wiener filtering for active listening of music from uncompressed (16-bit PCM) multichannel mix signals. In the present work, the system is improved to work with (MPEG-2 AAC) compressed mix signals: quantization noise is estimated from the AAC bitstream at the decoder and explicitly taken into account in the source separation process. Also a direct MDCT-to-STFT transform is used to optimize the computational efficiency of the process in the STFT domain from AAC-decoded MDCT coefficients.
Keywords :
Wiener filters; source separation; video coding; AAC bitstream; AAC decoded MDCT coefficients; MPEG-2 AAC compressed mix signals; STFT domain; Wiener filtering; compressed mixtures; computational efficiency; decoder; informed source separation system; music; quantization noise estimation; source separation process; spatial Wiener filter; uncompressed multichannel mix signals; Bit rate; Decoding; Encoding; Estimation; Noise; Noise reduction; Quantization (signal); AAC; Denoising; Informed Source Separation; NTF; Wiener Filter;
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2013 IEEE International Conference on
Conference_Location :
Vancouver, BC
DOI :
10.1109/ICASSP.2013.6637609
