Name
Title | ||
|---|---|---|
Speech Dereverberation Based on Maximum-Likelihood Estimation With Time-Varying Gaussian Source Model |
Abstract | ||
|---|---|---|
Distant acquisition of acoustic signals in an enclosed space often produces reverberant components due to acoustic reflections in the room. Speech dereverberation is in general desirable when the signal is acquired through distant microphones in such applications as hands-free speech recognition, teleconferencing, and meeting recording. This paper proposes a new speech dereverberation approach based on a statistical speech model. A time-varying Gaussian source model (TVGSM) is introduced as a model that represents the dynamic short time characteristics of nonreverberant speech segments, including the time and frequency structures of the speech spectrum. With this model, dereverberation of the speech signal is formulated as a maximum-likelihood (ML) problem based on multichannel linear prediction, in which the speech signal is recovered by transforming the observed signal into one that is probabilistically more like nonreverberant speech. We first present a general ML solution based on TVGSM, and derive several dereverberation algorithms based on various source models. Specifically, we present a source model consisting of a finite number of states, each of which is manifested by a short time speech spectrum, defined by a corresponding autocorrelation (AC) vector. The dereverberation algorithm based on this model involves a finite collection of spectral patterns that form a codebook. We confirm experimentally that both the time and frequency characteristics represented in the source models are very important for speech dereverberation, and that the prior knowledge represented by the codebook allows us to further improve the dereverberated speech quality. We also confirm that the quality of reverberant speech signals can be greatly improved in terms of the spectral shape and energy time-pattern distortions from simply a short speech signal using a speaker-independent codebook. |
Year | DOI | Venue |
|---|---|---|
2008 | 10.1109/TASL.2008.2004306 | IEEE Transactions on Audio, Speech & Language Processing |
Keywords | Field | DocType |
Gaussian processes,acoustic correlation,architectural acoustics,maximum likelihood estimation,probability,reverberation,source separation,spectral analysis,speech processing,acoustic reflection,acoustic signal,autocorrelation vector,dynamic short time characteristics,maximum-likelihood estimation,multichannel linear prediction,nonreverberant speech segment,probability,reverberant speech signal quality,short-time speech spectrum,source model,speaker-independent codebook,spectral pattern,speech dereverberation algorithm,speech quality,statistical speech model,time-pattern distortion,time-varying Gaussian source model,Blind signal processing,dereverberation,maximum-likelihood (ML) estimation,multichannel linear prediction,speech,time-varying Gaussian source model (TVGSM) | Speech processing,Speech coding,Pattern recognition,Computer science,Speech recognition,Linear prediction,Artificial intelligence,Blind signal separation,Source separation,Linear predictive coding,Codebook,Acoustic model | Journal |
Volume | Issue | ISSN |
16 | 8 | 1558-7916 |
Citations | PageRank | References |
14 | 0.93 | 20 |
Authors | ||
6 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Tomohiro Nakatani | 1 | 1327 | 139.18 |
| Biing-Hwang Juang | 2 | 3388 | 699.72 |
| T. Yoshioka | 3 | 112 | 4.62 |
| K. Kinoshita | 4 | 88 | 3.54 |
| Marc Delcroix | 5 | 699 | 62.07 |
| Masato Miyoshi | 6 | 709 | 74.28 |