Paper Info
Title
Joint Optimization Of Recurrent Networks Exploiting Source Auto-Regression For Source Separation
Abstract
In music interferences condition, source separation is very difficult. In this paper, we propose a novel recurrent network exploiting the auto-regressions of speech and music interference for source separation. An auto-regression can capture the shortterm temporal dependencies in data to help the source separation. For the separation, we independently separate the magnitude spectra of speech and interference from the mixture spectra by including an extra masking layer in the recurrent network. Compared to directly evaluating the ideal mask, the extra masking layer relaxes the assumption of independence between speech and interference which is more suitable for the real world environments. Using the separated spectra of speech and interference, we further explore a discriminative training objective and joint optimization framework for the proposed network, which incorporates the correlations and spectral dependencies of speech and interference into the separation. Systematic experiments show that the proposed model is competitive with the state-of-the-art method in singing-voice separations.
Year
Venue
Keywords
2015
16TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2015), VOLS 1-5
source separation, deep recurrent neural networks, discriminative training objective, autoregressive models
Field
DocType
Citations 
Autoregressive model,Pattern recognition,Computer science,Speech recognition,Artificial intelligence,Source separation
Conference
2
PageRank 
References 
Authors
0.35
9
7
Name
Order
Citations
PageRank
Shuai Nie1408.30
Wei Xue231.39
Shan Liang3208.52
Xueliang Zhang48019.41
Wenju Liu521439.32
Liwei Qiao630.73
Jianping Li720.35