Multitaper Estimation of Frequency-Warped Cepstra with Application to Speaker Verification
Publikation/Tidskrift/Serie: IEEE Signal Processing Letters
Dokumenttyp: Artikel i tidskrift
Förlag: IEEE--Institute of Electrical and Electronics Engineers Inc.
Usually the mel-frequency cepstral coefficients are estimated either from a periodogram or from a windowed periodogram. We state a general estimator which also includes multitaper estimators. We propose approximations of the variance and bias of the estimate of each coefficient. By using Monte Carlo computations, we demonstrate that the approximations are accurate. Using the proposed formulas, the peak matched multitaper estimator is shown to have low mean square error (squared bias + variance) on speech-like processes. It is also shown to perform slightly better in the NIST 2006 speaker verification task as compared to the Hamming window conventionally used in this context.
- Probability Theory and Statistics
- Speech analysis
- Speaker verification
- Cepstral analysis
- Multiple windows
- Statistical Signal Processing Group
- ISSN: 1070-9908