We present spectral excitation coding (SEC), a speech codec based on a sinusoidal model applied to the excitation signal. A phase dispersion algorithm allows the same model to be used for voiced as well as unvoiced and transitional sounds. The phase dispersion algorithm significantly improves the perceived quality resulting in more natural reconstructed speech. A new technique for variable dimension vector quantization called nonsquare transform vector quantization (NSTVQ) is used for quantization of the harmonic magnitudes. The SEC system at 2.45 kb/s achieved an MOS score 0.8 points higher than the 2.4 kb/s ZPC-10 standard. A preliminary 1.85 kb/s SEC system which uses zero-bit magnitude quantization is also presented. Informal listening tests indicate that the quality of the 1.85 kb/s system exceeds that of the LPC-10 standard.

A tree-searched multistage vector quantization (VQ) scheme for linear prediction coding (LPC) parameters which achieves spectral distortion lower than 1 dB with low complexity and good robustness using rates as low as 22 b/frame is presented. The M-L search is used, and it is shown that it achieves performance close to that of the optimal search for a relatively small M. A joint codebook design strategy for multistage VQ which improves convergence speed and the VQ performance measures is presented. The best performance/complexity tradeoffs are obtained with relatively small size codebooks cascaded in a 3-6 stage configuration. It is shown experimentally that as the number of stages is increased above the optimal performance/complexity tradeoff, the quantizer robustness and outlier performance can be improved at the expense of a slight increase in rate. Results for log area ratio (LAR) and line spectral pairs (LSPs) parameters are presented. A training technique that reduces outliers at the expense of a slight average performance degradation is introduced. The method significantly outperforms the split codebook approach.>