A 1.44 kbps vocoder based on LSP

This paper presents a simple and efficient speech coding algorithm based on traditional LPC vocoder. By employing a dynamic smoothing method to optimize the pitch detector and split vector quantization to quantize the LSP parameters, a high level of speech quality and intelligibility is achieved at 1.44 kbps. Because of the low computational complexity and high coding efficiency, it is one of the best algorithms suitable for ASIC design.

[1]  Kuldip K. Paliwal,et al.  Efficient vector quantization of LPC parameters at 24 bits/frame , 1990, [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing.

[2]  Frank K. Soong,et al.  Optimal quantization of LSP parameters using delayed decisions , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[3]  Nariman Farvardin,et al.  Quantizer design in LSP speech analysis-synthesis , 1988, IEEE J. Sel. Areas Commun..

[4]  Kuldip K. Paliwal,et al.  Efficient vector quantization of LPC parameters at 24 bits/frame , 1993, IEEE Trans. Speech Audio Process..

[5]  Nariman Farvardin,et al.  Finite-state vector quantization over noisy channels and its application to LSP parameters , 1992, [Proceedings] ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[6]  Biing-Hwang Juang,et al.  Line spectrum pair (LSP) and speech data compression , 1984, ICASSP.

[7]  Takehiro Moriya,et al.  Coding of LSP Parameters Using Interframe Moving Average Prediction and Multi-Stage Vector Quantization , 1993, Proceedings., IEEE Workshop on Speech Coding for Telecommunications,.

[8]  Nariman Farvardin,et al.  Coding of speech LSP parameters using TSVQ with interblock noiseless coding , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[9]  Rajiv Laroia,et al.  Efficient encoding of speech LSP parameters using the discrete cosine transformation , 1989, International Conference on Acoustics, Speech, and Signal Processing,.