Page 13 - 《应该声学》2022年第2期

第 41 卷 第 2 期 康坊等： 子带 t 分布的快速独立向量分析在语声盲源分离中的应用 181

[10] Liang Y, Naqvi S M, Chambers J. Overcoming block per- [13] Nakamura S, Hiyane K, Asano F, et al. Acous-
mutation problem in frequency domain blind source sepa- tical sound database in real environments for sound
ration when using AuxIVA algorithm[J]. Electronics Let- scene understanding and hands-free speech recogni-
ters, 2012, 48(8): 460–462. tion[C]//International Conference on Language Resources
[11] Scheibler R, Ono N. Fast and stable blind source sepa- and Evaluation (LREC), 2000: 965–968.
ration with Rank-1 updates[C]//2020 IEEE International [14] Garofolo J S, Lamel L F, Fisher W M, et al. DARPA
Conference on Acoustics, Speech and Signal Processing TIMIT acoustic-phonetic continous speech corpus CD-
(ICASSP). IEEE, 2020: 236–240. ROM. NIST speech disc 1–1.1[R]. NASA STI/Recon Tech-
[12] Harris J, Rivet B, Naqvi S M, et al. Real-time inde- nical Report N, 1993.
pendent vector analysis with student’s t source prior for [15] Vincent E, Gribonval R, Févotte C. Performance mea-
convolutive speech mixtures[C]//2015 IEEE International surement in blind audio source separation[J]. IEEE Trans-
Conference on Acoustics, Speech and Signal Processing actions on Audio, Speech & Language Processing, 2006,
(ICASSP). IEEE, 2015: 1856–1860. 14(4): 1462–1469.