Speech coding based upon vector quantization
Andres Buzo, A. Gray, Robert M. Gray et al. · IEEE Transactions on Acoustics Speech and Signal Processing · 1980
With rare exception, all presently available narrow-band speech coding systems implement scalar quantization (independent quantization) of the transmission parameters (such as reflection coefficients or transformed reflection coefficients in LPC systems). This paper presents a new approach called vector quantization. For very low data rates, realistic experiments have shown that vector quantization can achieve a given level of average distortion with 15 to 20 fewer bits/frame than that required for the optimized scalar quantizing approaches presently in use. The vector quantizing approach is shown to be a mathematically and computationally tractable method which builds upon knowledge obtained in linear prediction analysis studies. This paper introduces the theory in a nonrigorous form, along with practical results to date and an extensive list of research topics for this new area of speech coding.
Cited by 559 works. doi:10.1109/tassp.1980.1163445
Authors
References
- An Algorithm for Vector Quantizer Design
- Least squares quantization in PCM
- Linear Prediction of Speech
- Linear prediction: A tutorial review
- Distortion measures for speech processing
- Distance measures for speech processing
- Quantization properties of transmission parameters in linear predictive systems
- Comparison of optimal quantizations of speech reflection coefficients
- Rate-distortion speech coding with a minimum discrimination information distortion measure
- Quantization and bit allocation in speech processing