Abstract
An iterative training method is proposed for the design of random codebooks and channel code mapping tables for a CELP coder. The method is based on a structured excitation source, including a conjugate structure, pitch period repetition, and integrated mapping to channel code. All training procedures use a closed-loop distortion measure, which minimize perceptually weighted distortion between the input and the synthesized speech. The training improves SNR by around 1 dB and improves subjective quality. The channel code mapping and channel-matched distortion measure increase robustness against channel errors.
| Original language | English |
|---|---|
| Pages (from-to) | 397-404 |
| Number of pages | 8 |
| Journal | NTT R and D |
| Volume | 43 |
| Issue number | 4 |
| Publication status | Published - 1994 |
| Externally published | Yes |
ASJC Scopus subject areas
- Electrical and Electronic Engineering
Cite this
Training algorithm of the codebooks. / Moriya, Takehiro; Miki, Satoshi; Mano, Kazunori; Ikedo, Jotarou.
In: NTT R and D, Vol. 43, No. 4, 1994, p. 397-404.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - Training algorithm of the codebooks
AU - Moriya, Takehiro
AU - Miki, Satoshi
AU - Mano, Kazunori
AU - Ikedo, Jotarou
PY - 1994
Y1 - 1994
N2 - An iterative training method is proposed for the design of random codebooks and channel code mapping tables for a CELP coder. The method is based on a structured excitation source, including a conjugate structure, pitch period repetition, and integrated mapping to channel code. All training procedures use a closed-loop distortion measure, which minimize perceptually weighted distortion between the input and the synthesized speech. The training improves SNR by around 1 dB and improves subjective quality. The channel code mapping and channel-matched distortion measure increase robustness against channel errors.
AB - An iterative training method is proposed for the design of random codebooks and channel code mapping tables for a CELP coder. The method is based on a structured excitation source, including a conjugate structure, pitch period repetition, and integrated mapping to channel code. All training procedures use a closed-loop distortion measure, which minimize perceptually weighted distortion between the input and the synthesized speech. The training improves SNR by around 1 dB and improves subjective quality. The channel code mapping and channel-matched distortion measure increase robustness against channel errors.
UR - http://www.scopus.com/inward/record.url?scp=0028134444&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0028134444&partnerID=8YFLogxK
M3 - Article
VL - 43
SP - 397
EP - 404
JO - NTT R and D
JF - NTT R and D
SN - 0915-2326
IS - 4
ER -