Abstract
A transform coding of speech that uses a weighted vector quantization scheme is propopsed. First, the characteristics of the weighted vector quantization are checked. Then based on these considerations, a frequency-domain coder is designed for medium-band (4. 8-9. 6 kb/s) speech coding. In this coding scheme, the linear prediction residue is transformed and vector-quantized. In order to control the quantization errors in the frequency domain, adaptively weighted matching is used instead of the conventional adaptive bit allocation. Therefore, the residual signal can be reconstructed by the decoder, even if the spectral envelope parameters are destroyed due to transmission errors. The coded speech is natural and unaffected by background noise and its mean opinion score at 7. 2 kb/ps is comparable to that of 5. 5-b log PCM-coded speech.
| Original language | English |
|---|---|
| Title of host publication | ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings |
| Publisher | IEEE |
| Pages | 1629-1632 |
| Number of pages | 4 |
| Publication status | Published - 1987 |
| Externally published | Yes |
Fingerprint
ASJC Scopus subject areas
- Signal Processing
- Electrical and Electronic Engineering
- Acoustics and Ultrasonics
Cite this
TRANSFORM CODING OF SPEECH WITH WEIGHTED VECTOR QUANTIZATION. / Moriya, Takehiro; Honda, Masaaki.
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings. IEEE, 1987. p. 1629-1632.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - TRANSFORM CODING OF SPEECH WITH WEIGHTED VECTOR QUANTIZATION.
AU - Moriya, Takehiro
AU - Honda, Masaaki
PY - 1987
Y1 - 1987
N2 - A transform coding of speech that uses a weighted vector quantization scheme is propopsed. First, the characteristics of the weighted vector quantization are checked. Then based on these considerations, a frequency-domain coder is designed for medium-band (4. 8-9. 6 kb/s) speech coding. In this coding scheme, the linear prediction residue is transformed and vector-quantized. In order to control the quantization errors in the frequency domain, adaptively weighted matching is used instead of the conventional adaptive bit allocation. Therefore, the residual signal can be reconstructed by the decoder, even if the spectral envelope parameters are destroyed due to transmission errors. The coded speech is natural and unaffected by background noise and its mean opinion score at 7. 2 kb/ps is comparable to that of 5. 5-b log PCM-coded speech.
AB - A transform coding of speech that uses a weighted vector quantization scheme is propopsed. First, the characteristics of the weighted vector quantization are checked. Then based on these considerations, a frequency-domain coder is designed for medium-band (4. 8-9. 6 kb/s) speech coding. In this coding scheme, the linear prediction residue is transformed and vector-quantized. In order to control the quantization errors in the frequency domain, adaptively weighted matching is used instead of the conventional adaptive bit allocation. Therefore, the residual signal can be reconstructed by the decoder, even if the spectral envelope parameters are destroyed due to transmission errors. The coded speech is natural and unaffected by background noise and its mean opinion score at 7. 2 kb/ps is comparable to that of 5. 5-b log PCM-coded speech.
UR - http://www.scopus.com/inward/record.url?scp=0023167182&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0023167182&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:0023167182
SP - 1629
EP - 1632
BT - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
PB - IEEE
ER -