Name
Abstract | ||
|---|---|---|
We propose novel approaches for equalizing the modulation spectrum for robust feature extraction in speech recognition. Common to all approaches in that the temporal trajectories of the feature parameters are first transformed into the magnitude modulation spectrum. In spectral histogram equalization (SHE) and two-band spectral histogram equalization (2B-SHE), we equalize the histogram of the modulation spectrum for each utterance to a reference histogram obtained from clean training data, or perform the equalization with two sub-bands on the modulation spectrum. In magnitude ratio equalization (MRE), we define the magnitude ratio of lower to higher modulation frequency components for each utterance, and equalize this to a reference value obtained from clean training data. These approaches can be viewed as temporal filters that are adapted to each testing utterance. Experiments performed on the Aurora 2 and 4 corpora for small and large vocabulary tasks indicate that significant performance improvements are achievable for all noise conditions. We also show that additional improvements can be obtained when these approaches are integrated with cepstral mean and variance normalization (CMVN), histogram equalization (HEQ), higher order cepstral moment normalization (HOCMN), or the advanced front-end (AFE). We analyze and discuss the reasons for these improvements from different viewpoints with different sets of data, including adaptive temporal filtering, noise behavior on the modulation spectrum, phoneme types, and modulation spectrum distance measures. |
Year | DOI | Venue |
|---|---|---|
2012 | 10.1109/TASL.2011.2166544 | IEEE Transactions on Audio, Speech & Language Processing |
Keywords | Field | DocType |
modulation spectrum equalization,two-band spectral histogram equalization,speech recognition,spectral histogram equalization,reference value,mre,adaptive temporal filtering,modulation spectrum,higher order cepstral moment normalization,clean training data,feature normalization,cmvn,can clean training data,improved robust speech recognition,afe,heq,adaptive filters,feature extraction,advanced front-end,reference histogram,temporal filter,magnitude ratio equalization,modulation spectrum distance measure,magnitude modulation spectrum,cepstral mean and variance normalization,vocabulary tasks,higher modulation frequency component,modulation frequency components,2b-she,histogram equalization,robust feature extraction,hocmn,wiener filter,higher order,difference set,signal to noise ratio,spectrum,band pass filter,modulation,front end,band pass filters,histograms | Histogram,Normalization (statistics),Equalization (audio),Pattern recognition,Computer science,Signal-to-noise ratio,Adaptive histogram equalization,Speech recognition,Cepstral Mean and Variance Normalization,Artificial intelligence,Frequency modulation,Histogram equalization | Journal |
Volume | Issue | ISSN |
20 | 3 | 1558-7916 |
Citations | PageRank | References |
14 | 0.75 | 36 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Liang-Che Sun | 1 | 36 | 3.43 |
| Lin-shan Lee | 2 | 1525 | 182.03 |