摘要
关键词识别是语音识别中一个重要的研究方向。该文提出了一种基于HMM模型的关键词识别方法。研究并实现了一种新的具有无废料模型精度的采用迭代viterbi的解码算法,从而提高了关键词系统的识别效率。
Keyword spotting is one of important topics in speech recognition. This paper presents an approach of hidden markov model(HMM)-based keyword spotting, researches and implements an iterative viterbi decoding method which has the precision of non-filler models and improves the efficiency of the system.
出处
《计算机工程》
CAS
CSCD
北大核心
2004年第7期130-132,共3页
Computer Engineering