Discrete Time Processing of Speech Signals
Discrete Time Processing of Speech Signals
The convolution theorem for the continuous wavelet tranform
Signal Processing
| Hi-index | 0.00 |
Cepstral analysis has been used on voiced speech to separate (deconvolve) the vocal tract filtering effect from the excitation produced by vocal fold vibration (voicing). This paper presents a new approach to speech deconvolution via the biorthogonal wavelet decomposition and reconstruction. The results of some experiments using wavelet deconvolution with voiced speech are given, and these results are compared with the cepstral method. They show that the wavelet method has the property of robustness. It is also automatic and easy to implement.