ONLINE IVA WITH ADAPTIVE LEARNING FOR SPEECH SEPARATION USING VARIOUS SOURCE PRIORS

被引:0
作者
Erateb, Suleiman [1 ]
Naqvi, Mohsen [2 ]
Chambers, Jonathon [2 ]
机构
[1] Loughborough Univ Technol, Wolfson Sch Mech Mfg & Elect Engn, Loughborough LE11 3TU, Leics, England
[2] Newcastle Univ, Sch Elect & Elect Engn, Newcastle Upon Tyne NE1 7RU, Tyne & Wear, England
来源
2017 SENSOR SIGNAL PROCESSING FOR DEFENCE CONFERENCE (SSPD) | 2017年
关键词
Blind source separation; convolutive mixture; independent vector analysis; online; adaptive learning; room impulse responses; INDEPENDENT VECTOR ANALYSIS; BLIND SOURCE SEPARATION;
D O I
暂无
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Independent vector analysis (IVA) is a frequency domain blind source separation (FDBSS) technique that has proven efficient in separating independent speech signals from their convolutive mixtures. In particular, it addresses the problematic permutation problem by using a multivariate source prior. The multivariate source prior models statistical inter dependency across the frequency bins of each source and the performance of the method is dependent upon the choice of source prior. The online form of the IVA is suitable for practical real time systems. Previous online algorithms use a learning rate that does not introduce a robust way to control the learning as a function of the proximity to the target solution. In this work, we propose a new adaptive learning scheme to improve the convergence speed and steady state separation performance. The speech signals are modelled by two different source priors; a super-Gaussian distribution and a generalized Gaussian distribution. The experimental results confirm improved performance with real room impulse responses and real recorded speech signals.
引用
收藏
页码:74 / 78
页数:5
相关论文
共 17 条
[1]   IMAGE METHOD FOR EFFICIENTLY SIMULATING SMALL-ROOM ACOUSTICS [J].
ALLEN, JB ;
BERKLEY, DA .
JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA, 1979, 65 (04) :943-950
[2]  
Amari S., 1996, ADV NEURAL INFORMATI, V8, P752
[3]  
[Anonymous], 1993, 4930 NISTIR
[4]  
[Anonymous], 2007, MULTICHANNEL SPEECH
[5]  
Bingham E, 2000, Int J Neural Syst, V10, P1, DOI 10.1142/S0129065700000028
[6]   SOME FURTHER EXPERIMENTS UPON THE RECOGNITION OF SPEECH, WITH ONE AND WITH 2 EARS [J].
CHERRY, EC ;
TAYLOR, WK .
JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA, 1954, 26 (04) :554-559
[7]  
Das N., 2007, NEURAL INFORM PROCES, V11, P225
[8]  
Harris J, 2015, INT CONF ACOUST SPEE, P1856, DOI 10.1109/ICASSP.2015.7178292
[9]   Blind source separation exploiting higher-order frequency dependencies [J].
Kim, Taesu ;
Attias, Hagai T. ;
Lee, Soo-Young ;
Lee, Te-Won .
IEEE TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2007, 15 (01) :70-79
[10]   Real-Time Independent Vector Analysis for Convolutive Blind Source Separation [J].
Kim, Taesu .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS, 2010, 57 (07) :1431-1438