Quantization Issues in Harmonic Coders

For some years now, vector quantization has been used as a synonym of efficient quantization schemes, despite the large computational overhead involved in this type of approaches. The introduction of vector quantization in the harmonic coder framework may affect several sets of parameters, namely the amplitudes and the phases of the harmonic coefficients. This paper reports a systematic experimental study, in which this type of quantization is compared with traditional scalar techniques, for quantizing the harmonic parameters. The experiments showed that it is possible to quantize voiced speech below 8 kbps, and still obtain high quality synthetic signals.

[1]  R. McAulay,et al.  Mid-rate coding based on a sinusoidal representation of speech , 1985, ICASSP '85. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[2]  Manfred R. Schroeder,et al.  Code-excited linear prediction(CELP): High-quality speech at very low bit rates , 1985, ICASSP '85. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[3]  E. Bronson,et al.  Harmonic coding of speech at 4.8 kb/s , 1987, ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[4]  R. Gray,et al.  Speech coding based upon vector quantization , 1980, ICASSP.

[5]  Ronald E. Crochiere,et al.  Frequency domain coding of speech , 1979 .

[6]  Luís B. Almeida,et al.  New basis functions for sinusoidal decompositions , 1988, 8th European Conference on Electrotechnics, Conference Proceedings on Area Communication.

[7]  Luís B. Almeida,et al.  Variable-frequency synthesis: An improved harmonic coding scheme , 1984, ICASSP.