基于连续语音识别算法和词树约束的汉语词组语音识别

具有确定词表的词组语音识别是语音识别研究的一个重要方面,应用相当广泛。本文在简单介绍词组语音识别之后给出了一个基于连续语音识别算法和词树约束的汉语词组语音识别方法。这种方法通过在束搜索连续语音识别算法中引入词树约束信息,发挥了连续语音识别算法的优点,并且充分利用了确定词表的约束信息,提高了计算和搜索的效率。然后介绍了约束词树和它的高效存储结构,这种结构提高了约束词树的存储效率和在识别搜索中的检索效率;最后给出实验的结果和讨论并进行简要的总结。

Chinese Phrase Recognition Based on Continuous Speech Recognition Algorithm and Phrase-Tree Constraint

Phrase recognition in fixed phrase table, with many applications, is an important research area of speech recognition. After an overview of phrase recognition, this paper proposes a Chinese phrase recognition method based on continuous speech recognition algorithm and phrase-tree constraint. This method gets high computing and searching efficiency by applying phrase-tree constraint information in continuous speech recognition algorithm. We first introduce the beamsearch continuous speech recognition algorithm. Then the restrictive phrase-tree is presented with its compact storage structure that improves the storage and access efficiency in recognition search. Finally, the experiment results and discussion are given followed by a brief summary.