Speech recognition is a technology that always seems two to 10 years away from speaker-independent recognition of a largevocabulary.
要实现与说话者不相关的大型词汇表识别，语言识别技术还要2到10年的路要走。
2
Largevocabulary continuous speech recognition is main direction of speech recognition research at present, in which the research on particularity of tone recognition is an important work.
大词汇连续语音识别是当前语音识别研究的主流，探讨其中声调识别的特殊性是一项有意义的工作。
3
In this paper, error tolerant techniques are studied for largevocabulary speaker independent Chinese continuous speech recognition and understanding systems.