The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
Neste artigo, é estudado o uso da transformada ótima de Karhunen-Loeve (KL) para quantização dos coeficientes de frequência do espectro da linha de fala (LSF). Os esquemas de quantizador escalar (SQ) e quantizador vetorial (VQ) são desenvolvidos para codificar eficientemente os parâmetros de transformação após operar a transformada KL unidimensional ou bidimensional. Além disso, os esquemas SQ também são combinados com codificação de entropia usando codificação de comprimento variável Huffman (VLC). A idéia básica no desenvolvimento desses esquemas é utilizar a forte correlação dos parâmetros LSF para reduzir a taxa de bits para um determinado nível de fidelidade. Como o uso de estatísticas globais para gerar o esquema de codificação pode não ser apropriado, propomos vários sistemas adaptativos de transformada KL (AKL) para codificar os parâmetros LSF. O desempenho de todos os sistemas para diferentes taxas de bits é investigado e comparações adequadas são feitas. É mostrado que os sistemas de codificação por transformada KL propostos apresentam um desempenho tão bom ou melhor para SQ e VQ nas taxas de bits examinadas em comparação com outros métodos no campo da codificação LSF.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Laszlo LOIS, Hai Le VU, "Spectral Coding of Speech LSF Parameters Using Karhunen-Loeve Transform" in IEICE TRANSACTIONS on Fundamentals,
vol. E82-A, no. 10, pp. 2138-2146, October 1999, doi: .
Abstract: In this paper, the use of optimal Karhunen-Loeve (KL) transform for quantization of speech line spectrum frequency (LSF) coefficients is studied. Both scalar quantizer (SQ) and vector quantizer (VQ) schemes are developed to encode efficiently the transform parameters after operating one or two-dimensional KL transform. Furthermore, the SQ schemes are also combined with entropy coding by using Huffman variable length coding (VLC). The basic idea in developing these schemes is utilizing the strong correlation of LSF parameters to reduce the bit rate for a given level of fidelity. Since the use of global statistics for generating the coding scheme may not be appropriate, we propose several adaptive KL transform systems (AKL) to encode the LSF parameters. The performance of all systems for different bit rates is investigated and adequate comparisons are made. It is shown that the proposed KL transform coding systems introduce as good as or better performance for both SQ and VQ in the examined bit rates compared to other methods in the field of LSF coding.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e82-a_10_2138/_p
Copiar
@ARTICLE{e82-a_10_2138,
author={Laszlo LOIS, Hai Le VU, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Spectral Coding of Speech LSF Parameters Using Karhunen-Loeve Transform},
year={1999},
volume={E82-A},
number={10},
pages={2138-2146},
abstract={In this paper, the use of optimal Karhunen-Loeve (KL) transform for quantization of speech line spectrum frequency (LSF) coefficients is studied. Both scalar quantizer (SQ) and vector quantizer (VQ) schemes are developed to encode efficiently the transform parameters after operating one or two-dimensional KL transform. Furthermore, the SQ schemes are also combined with entropy coding by using Huffman variable length coding (VLC). The basic idea in developing these schemes is utilizing the strong correlation of LSF parameters to reduce the bit rate for a given level of fidelity. Since the use of global statistics for generating the coding scheme may not be appropriate, we propose several adaptive KL transform systems (AKL) to encode the LSF parameters. The performance of all systems for different bit rates is investigated and adequate comparisons are made. It is shown that the proposed KL transform coding systems introduce as good as or better performance for both SQ and VQ in the examined bit rates compared to other methods in the field of LSF coding.},
keywords={},
doi={},
ISSN={},
month={October},}
Copiar
TY - JOUR
TI - Spectral Coding of Speech LSF Parameters Using Karhunen-Loeve Transform
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 2138
EP - 2146
AU - Laszlo LOIS
AU - Hai Le VU
PY - 1999
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E82-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 1999
AB - In this paper, the use of optimal Karhunen-Loeve (KL) transform for quantization of speech line spectrum frequency (LSF) coefficients is studied. Both scalar quantizer (SQ) and vector quantizer (VQ) schemes are developed to encode efficiently the transform parameters after operating one or two-dimensional KL transform. Furthermore, the SQ schemes are also combined with entropy coding by using Huffman variable length coding (VLC). The basic idea in developing these schemes is utilizing the strong correlation of LSF parameters to reduce the bit rate for a given level of fidelity. Since the use of global statistics for generating the coding scheme may not be appropriate, we propose several adaptive KL transform systems (AKL) to encode the LSF parameters. The performance of all systems for different bit rates is investigated and adequate comparisons are made. It is shown that the proposed KL transform coding systems introduce as good as or better performance for both SQ and VQ in the examined bit rates compared to other methods in the field of LSF coding.
ER -