Title :
Modulation Spectral Features for Robust Far-Field Speaker Identification
Author :
Falk, Tiago H. ; Chan, Wai-Yip
Author_Institution :
Dept. of Electr. & Comput. Eng., Queen´´s Univ., Kingston, ON, Canada
Abstract :
In this paper, auditory inspired modulation spectral features are used to improve automatic speaker identification (ASI) performance in the presence of room reverberation. The modulation spectral signal representation is obtained by first filtering the speech signal with a 23-channel gammatone filterbank. An eight-channel modulation filterbank is then applied to the temporal envelope of each gammatone filter output. Features are extracted from modulation frequency bands ranging from 3-15 H z and are shown to be robust to mismatch between training and testing conditions and to increasing reverberation levels. To demonstrate the gains obtained with the proposed features, experiments are performed with clean speech, artificially generated reverberant speech, and reverberant speech recorded in a meeting room. Simulation results show that a Gaussian mixture model based ASI system, trained on the proposed features, consistently outperforms a baseline system trained on mel-frequency cepstral coefficients. For multimicrophone ASI applications, three multichannel score combination and adaptive channel selection techniques are investigated and shown to further improve ASI performance.
Keywords :
Gaussian processes; filtering theory; modulation; signal representation; speaker recognition; speech processing; Gaussian mixture model; auditory inspired modulation; automatic speaker identification; gammatone filterbank; modulation filterbank; modulation spectral features; modulation spectral signal representation; robust far-field speaker identification; room reverberation; speech signal filtering; Gaussian mixture model (GMM); modulation spectrum; reverberation; reverberation time; speaker identification;
Journal_Title :
Audio, Speech, and Language Processing, IEEE Transactions on
DOI :
10.1109/TASL.2009.2023679