New feature parameters for speech recognition
碩士 === 輔仁大學 === 資訊工程學系 === 99 === Abstract Speech recognition technology mostly takes MFCC as feature vector, and uses HTK (HMM Toolkits) to train database and build acoustical model. In this thesis, we use neural network to train database and build acoustical model, and make our system recognition...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2010
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/01990713739835843691 |