Live speech recognition in sports games by adaptation of acoustic model and language model
Live speech recognition in sports games by adaptation of acoustic model and language model
复制标题
通过适应声学模型和语言模型进行体育比赛中的实时语音识别
DOI：
10.21437/eurospeech.2003-424
复制
发表时间：
2003
期刊：
影响因子：
--
通讯作者：
Masakiyo Fujimoto
中科院分区：
文献类型：
--
作者：
Y. Ariki;Takeru Shigemori;Tsuyoshi Kaneko;J. Ogata;Masakiyo Fujimoto
提出一种基于LVCSR的棒球广播语音关键词自动提取方法。为了鲁棒识别，我们采用了声学和语言模型自适应。在声学模型自适应中，使用MLLR+MAP进行有监督和无监督自适应。通过这两个级别的适应，单词准确率提高了28%。在语言模型自适应中，进行了语言模型融合和发音修改。这种适应在单词准确性方面提高了13%。最后，通过整合这两种适应，在单词准确率水平上提高了38%，在关键字准确率水平上提高了28%。
This paper proposes a method to automatically extract keywords from baseball radio speech through LVCSR for highlight scene retrieval. For robust recognition, we employed acoustic and language model adaptation. In acoustic model adaptation, supervised and unsupervised adaptations were carried out using MLLR+MAP. By this two level adaptation, word accuracy was improved by 28%. In language model adaptation, language model fusion and pronunciation modification were carried out. This adaptation showed 13% improvement at word accuracy. Finally, by integrating both adaptations, 38% improvement was achieved at word accuracy level and 28% improvement at keyword accuracy level.
影响因子：
4.3
作者：
LEGGETTER, CJ;WOODLAND, PC
通讯作者：
WOODLAND, PC