论文部分内容阅读
近年来大词汇量连续语音识别技术得到了迅速的发展,国内外研究机构加大了对汉语和英语语音识别技术的研究,然而,维吾尔语语音识别技术的研究工作最近才起步。建立了面向大词汇量的维吾尔语语音语料库,研究了维吾尔语声学模型和语言模型建模技术、解码技术,进行了面向大词汇量的维吾尔语连续语音识别实验。对维吾尔语大词汇量连续语音识别技术进一步发展中存在的问题进行了讨论。
In recent years, large vocabulary continuous speech recognition technology has been rapid development of domestic and foreign institutions to increase the Chinese and English speech recognition technology research, however, Uyghur language speech recognition technology research has only recently started. A Uyghur phonetic corpus for large vocabulary was established. Uyghur language speech model and speech model modeling technology and decoding technology were studied, and Uyghur continuous speech recognition experiments for large vocabulary were carried out. The existing problems in the further development of Uyghur Vocabulary Large Volume Consecutive Speech Recognition Technology are discussed.