Using the method of principle component to recognize the Mandarin word in the system of independent speaker
碩士 === 國立中興大學 === 統計學研究所 === 103 === This paper mainly discuss the 1391 mandarin words recognition using principal component analysis. The feature is Mel-frequency cepstrum coefficients(MFCC), first, in the voice database the consonant and the vowel frame are divided into one or three or five frame...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2015
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/64468823001757069938 |