Name
Abstract | ||
|---|---|---|
We address the classical problem of delta feature computation, and interpret the operation involved in terms of Savitzky-Golay (SG) filtering. Features such as the mel-frequency cepstral coefficients (MFCCs), obtained based on short-time spectra of the speech signal, are commonly used in speech recognition tasks. In order to incorporate the dynamics of speech, auxiliary delta and delta-delta features, which are computed as temporal derivatives of the original features, are used. Typically, the delta features are computed in a smooth fashion using local least-squares (LS) polynomial fitting on each feature vector component trajectory. In the light of the original work of Savitzky and Golay, and a recent article by Schafer in IEEE Signal Processing Magazine, we interpret the dynamic feature vector computation for arbitrary derivative orders as SG filtering with a fixed impulse response. This filtering equivalence brings in significantly lower latency with no loss in accuracy, as validated by results on a TIMIT phoneme recognition task. The SG filters involved in dynamic parameter computation can be viewed as modulation filters, proposed by Hermansky. |
Year | DOI | Venue |
|---|---|---|
2013 | 10.1109/LSP.2013.2244593 | Signal Processing Letters, IEEE |
Keywords | DocType | Volume |
filtering theory,least squares approximations,polynomial approximation,speech recognition,LS polynomial fitting,MFCC,Mel-frequency cepstral coefficients,SG filtering,Savitzky-Golay filtering,TIMIT phoneme recognition task,delta feature computation problem,dynamic feature vector computation,feature vector component trajectory,filtering equivalence,local least-square polynomial fitting,modulation filters,short-time spectra,speech recognition,speech signal,Dynamic features,Savitzky-Golay (SG) filtering,modulation filtering,speech recognition | Journal | 20 |
Issue | ISSN | Citations |
3 | 1070-9908 | 1 |
PageRank | References | Authors |
0.35 | 3 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Sunder Ram Krishnan | 1 | 1 | 0.35 |
| Mathew Magimai-Doss | 2 | 1 | 0.69 |
| Chandra Sekhar Seelamantula | 3 | 48 | 7.95 |