Paper Info
Title
Towards online source counting in speech mixtures applying a variational EM for complex Watson mixture models
Abstract
This contribution describes a step-wise source counting algorithm to determine the number of speakers in an offline sce-nario. Each speaker is identified by a variational expectation maximization (VEM) algorithm for complex Watson mixture models and therefore directly yields beamforming vectors for a subsequent speech separation process. An observation selection criterion is proposed which improves the robustness of the source counting in noise. The algorithm is compared to an alternative VEM approach with Gaussian mixture models based on directions of arrival and shown to deliver improved source counting accuracy. The article concludes by extending the offline algorithm towards a low-latency online estimation of the number of active sources from the streaming input data.
Year
DOI
Venue
2014
10.1109/IWAENC.2014.6954009
Acoustic Signal Enhancement
Keywords
Field
DocType
Gaussian processes,direction-of-arrival estimation,expectation-maximisation algorithm,mixture models,speech processing,variational techniques,Gaussian mixture models,beamforming vectors,complex Watson mixture models,directions of arrival estimation,low-latency online estimation,observation selection criterion,online source counting,speech mixtures,speech separation process,variational EM,variational expectation maximization algorithm,Bayes methods,Blind source separation,Directional statistics,Number of speakers,Speaker diarization
Beamforming,Pattern recognition,Expectation–maximization algorithm,Computer science,Robustness (computer science),Selection criterion,Artificial intelligence,Mixture model,Source separation,Separation process
Conference
Citations 
PageRank 
References 
2
0.41
6
Authors
4
Name
Order
Citations
PageRank
Lukas Drude19511.10
Aleksej Chinaev2223.05
Dang Hai Tran Vu3444.01
Reinhold Haeb-Umbach41487211.71