Abstract
A method is discussed for characterizing speaker individuality that could be used for speaker adaptation for speaker-independent phoneme recognition. To represent the individual traits of speakers, a speaker space is organized using distributions in a vector-quantized space. This space includes 200 speakers, each shown as a point in it. The experimental results for 200-speaker identification with 10 input words shows performance rates of 82% in first-order and 90% in the second-order case.
| Original language | English |
|---|---|
| Title of host publication | Proceedings - International Conference on Pattern Recognition |
| Place of Publication | New York, NY, USA |
| Publisher | IEEE |
| Pages | 1152-1154 |
| Number of pages | 3 |
| ISBN (Print) | 0818607424 |
| Publication status | Published - 1986 |
| Externally published | Yes |
Fingerprint
ASJC Scopus subject areas
- Engineering(all)
Cite this
ORGANIZATION OF SPEAKER SPACE BASED ON VECTOR QUANTIZATION METHOD. / Shirai, Katsuhiko; Mano, Kazunori; Ishige, Shunichi; Sano, Koji.
Proceedings - International Conference on Pattern Recognition. New York, NY, USA : IEEE, 1986. p. 1152-1154.Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
}
TY - GEN
T1 - ORGANIZATION OF SPEAKER SPACE BASED ON VECTOR QUANTIZATION METHOD.
AU - Shirai, Katsuhiko
AU - Mano, Kazunori
AU - Ishige, Shunichi
AU - Sano, Koji
PY - 1986
Y1 - 1986
N2 - A method is discussed for characterizing speaker individuality that could be used for speaker adaptation for speaker-independent phoneme recognition. To represent the individual traits of speakers, a speaker space is organized using distributions in a vector-quantized space. This space includes 200 speakers, each shown as a point in it. The experimental results for 200-speaker identification with 10 input words shows performance rates of 82% in first-order and 90% in the second-order case.
AB - A method is discussed for characterizing speaker individuality that could be used for speaker adaptation for speaker-independent phoneme recognition. To represent the individual traits of speakers, a speaker space is organized using distributions in a vector-quantized space. This space includes 200 speakers, each shown as a point in it. The experimental results for 200-speaker identification with 10 input words shows performance rates of 82% in first-order and 90% in the second-order case.
UR - http://www.scopus.com/inward/record.url?scp=0022985602&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=0022985602&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:0022985602
SN - 0818607424
SP - 1152
EP - 1154
BT - Proceedings - International Conference on Pattern Recognition
PB - IEEE
CY - New York, NY, USA
ER -