Ecole d'ingénieur et centre de recherche en Sciences du numérique

Blind audio source separation using Short+Long Term AR source models and spectrum matching

Schutz, Antony; Slock, Dirk T M

DSP/SPE 2011, 14th IEEE Digital Signal Processing & 6th Signal Processing Education Workshop, January 4-7, 2011, Sedona, Arizona, USA

Blind audio source separation (BASS) arises in a number of applications in speech and music processing such as speech enhancement, speaker diarization, automated music transcription etc. Generally, BASS methods consider multichannel signal capture. The single microphone case is the most difficult underdetermined case, but it often arises in practice. In the approach considered here, the main source identifiability comes from exploiting the presumed quasi-periodic nature of the sources via long-term autoregressive (AR) modeling. Indeed, musical note signals are quasi-periodic and so is voiced speech, which constitutes the most energetic part of speech signals. We furthermore exploit (e.g. speaker or instrument related) prior information in the spectral envelope of the source signals via short-term AR modeling. We present an iterative method based on the minimization of the (weighted) Itakura-Saito distance for estimating the source parameters directly from the mixture using frame based processing.

Document Doi Bibtex

Titre:Blind audio source separation using Short+Long Term AR source models and spectrum matching
Type:Conférence
Langue:English
Ville:Sedona
Pays:ÉTATS-UNIS
Date:
Département:Systèmes de Communication
Eurecom ref:3316
Copyright: © 2011 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
Bibtex: @inproceedings{EURECOM+3316, doi = {http://dx.doi.org/10.1109/DSP-SPE.2011.5739196 }, year = {2011}, title = {{B}lind audio source separation using {S}hort+{L}ong {T}erm {AR} source models and spectrum matching}, author = {{S}chutz, {A}ntony and {S}lock, {D}irk {T} {M} }, booktitle = {{DSP}/{SPE} 2011, 14th {IEEE} {D}igital {S}ignal {P}rocessing \& 6th {S}ignal {P}rocessing {E}ducation {W}orkshop, {J}anuary 4-7, 2011, {S}edona, {A}rizona, {USA} }, address = {{S}edona, {\'{E}}{TATS}-{UNIS}}, month = {01}, url = {http://www.eurecom.fr/publication/3316} }
Voir aussi: