Variable-dimension quantization of sinusoidal amplitudes using Gaussian mixture models
- 28 September 2004
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 1 (15206149) , 153
- https://doi.org/10.1109/icassp.2004.1325945
Abstract
In this paper, Gaussian mixture (GM) models are used to design variable-dimension quantizers according to a weighted distortion criterion. A general method for combining a variable-to-fixed dimension transform, with GM modeling and quantization, is proposed. The method provides a convenient and efficient way to encode the amplitudes in a sinusoidal speech coder. Quantizers designed according to the proposed scheme are evaluated both according to weighted distortion criteria, and with respect to a high-rate bound approximation of the distortion. Informal listening tests suggest that the amplitudes can be encoded without subjective loss in a wideband harmonic coder, at a rate around 40 bits per frame (for the amplitudes only).Keywords
This publication has 7 references indexed in Scilit:
- PDF optimized parametric vector quantization of speech line spectral frequenciesIEEE Transactions on Speech and Audio Processing, 2003
- Variable-dimension vector quantization of speech spectra for low-rate vocodersPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Coding of variable dimension speech spectral vectors using weighted nonsquare transform vector quantizationIEEE Transactions on Speech and Audio Processing, 2001
- Vector quantization based on Gaussian mixture modelsIEEE Transactions on Speech and Audio Processing, 2000
- Robust text-independent speaker identification using Gaussian mixture speaker modelsIEEE Transactions on Speech and Audio Processing, 1995
- Vector quantized MBE with simplified V/UV division at 3.0 kbit/sPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1993
- Vector Quantization and Signal CompressionPublished by Springer Nature ,1992