DocumentCode
3021494
Title
On the use of LSF intermodel interlacing property for spectral quantization
Author
Lee, Mi Suk ; Kim, Hong Kook ; Choi, Seung Ho ; Lee, Hwang Soo
Author_Institution
Dept. of Electr. Eng., Korea Adv. Inst. of Sci. & Technol., Taejon, South Korea
fYear
1999
fDate
1999
Firstpage
43
Lastpage
45
Abstract
The line spectral frequencies (LSFs) extracted from successive analysis orders are interlaced with each other. This intermodel interlacing property gives a new relationship between the closeness of LSFs and their spectral sensitivities, which enables us to propose a weighting function for LSF distortion measurement. By applying the proposed weighting function to an LSF quantizer, we can achieve better performance than when using the conventional heuristic functions. Moreover, the complexity of the proposed weighting function is much lower than that of the optimal weighting function, while their performances are almost the same
Keywords
linear predictive coding; quantisation (signal); spectral analysis; speech coding; LPC analysis polynomial; LSF distortion measurement; LSF intermodel interlacing property; formant bounded weighting function; heuristic functions; line spectral frequencies; optimal weighting function; performance; spectral quantization; spectral sensitivities; speech coding; successive analysis orders; Covariance matrix; Distortion measurement; Euclidean distance; Frequency; Humans; Linear predictive coding; Polynomials; Quantization; Reflection; Weight measurement;
fLanguage
English
Publisher
ieee
Conference_Titel
Speech Coding Proceedings, 1999 IEEE Workshop on
Conference_Location
Porvoo
Print_ISBN
0-7803-5651-9
Type
conf
DOI
10.1109/SCFT.1999.781478
Filename
781478
Link To Document