Generalized Cepstral Modeling of Degraded Speech and Its Application to Speech Enhancement (Special Section of Papers Selected from the 7th Digital Signal Processing Symymposium
スポンサーリンク
概要
- 論文の詳細を見る
This paper proposes a technique for estimating speech parameters in noisy environment. The technique uses a spectral model represented by generalized cepstrum and estimates the generalized cepstral coefficients from the speech which has been degraded by additive background noise. Parameter estimation is based on maximum a posteriori (MAP) estimation procedure. An iterative approach which has been formulated for all-pole modeling is applied to the generalized cepstral modeling. Generalized cepstral coefficients are obtained by an iterative procedure that consists of the unbiased estimation of log spectrum and noncausal Wiener filtering. Since the generalized cepstral model includes the all-pole model as a special case, the technique can be viewed as a generalization of the all-pole modeling based on MAP estimation. The proposed technique is applied to the enhancement of speech and several experimental results are also shown.
- 社団法人電子情報通信学会の論文
- 1993-08-25
著者
-
Imai Satoshi
Precision And Intelligence Labolatory Tokyo Institute Of Technology
-
Kanno Toshio
Precision and Intelligence Laboratory, Tokyo Institute of Technology
-
Kobayashi Takao
Precision and Intelligence Laboratory, Tokyo Institute of Technology
-
Kanno Toshio
Precision And Intelligence Laboratory Tokyo Institute Of Technology
-
Kobayashi Takao
Precision And Intelligence Laboratory Tokyo Institute Of Technology
関連論文
- Multisegment Multiple VQ Codebooks-Based Speaker Independent Isolated-Word Recognition Using Unbiased Mel Cepstrum
- Generalized Cepstral Modeling of Degraded Speech and Its Application to Speech Enhancement (Special Section of Papers Selected from the 7th Digital Signal Processing Symymposium
- Combining Multiple Classifiers in a Hybrid System for High Performance Chinese Syllable Recognition
- Harmonics Estimation Based on Instantaneous Frequency and Its Application to Pitch Determination of Speech
- A New Approach of Parsing and Search Based on the Divide and Conquer Strategy for Continuous Speech Recognition