Leveraging Proximity Cues and Concept Information for Language Model Adaptation in Speech Recognition

碩士 === 國立臺灣師範大學 === 資訊工程學系 === 102 === This thesis investigates and develops language model adaptation techniques for Mandarin large vocabulary continuous speech recognition (LVCSR) and its main contribution is two-fold. First, the so-called “bag-of-words” assumption of conventional topic models is...

Full description

Bibliographic Details
Main Author: 郝柏翰
Other Authors: 陳柏琳
Format: Others
Language:zh-TW
Published: 2014
Online Access:http://ndltd.ncl.edu.tw/handle/60754768721537830713