Language Modeling for Mixed Language Speech Recognition using Weighted Phrase Extraction

被引:0
作者
Li, Ying [1 ]
Fung, Pascale [1 ]
机构
[1] Hong Kong Univ Sci & Technol, Dept Elect & Comp Engn, Hong Kong, Hong Kong, Peoples R China
来源
14TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2013), VOLS 1-5 | 2013年
关键词
mixed language; language model; code switching;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
To train a code switching language model for mixed language speech recognition, we propose to assign weights to the sentence pairs in the parallel text data. The code switching language model which is composed of the code switching boundary prediction model, code switching translation model and reconstruction model is incorporated with a language for mixed language speech recognition. The code switching translation model which is trained using selected subsets of the sentence pairs in the parallel text data allows the decoder to make the decision whether a phrase is in the matrix language or in the embedded language. Moreover, we propose a weighting procedure while training the code switching translation model. We evaluate our methods on Mandarin-English code switching lecture speech and lunch conversations. Our proposed method reduces word error rate by a statistically significant 1.74% on the lecture speech, and by 1.29% on the lunch conversation over the conventional interpolated language model.
引用
收藏
页码:2598 / 2602
页数:5
相关论文
共 19 条
[1]  
[Anonymous], 9 ANN C INT SPEECH C
[2]  
Axelrod Amittai, P 2011 C EMP METH NA, P355
[3]  
Chan J.Y.C., CHIN SPOK LANG PROC, P293
[4]  
Coulmas F., 1998, The Handbook of Sociolinguistics
[5]  
Gumperz J., 1982, Discourse strategies, DOI 10.1017/CBO9780511611834
[6]  
Houwei Cao, 2010, Proceedings 7th International Symposium on Chinese Spoken Language Processing (ISCSLP 2010), P246, DOI 10.1109/ISCSLP.2010.5684900
[7]  
Li Y., 2011, ICASSP
[8]  
Li Y., 2013, ICASSP
[9]  
MacSwan, 2012, HDB BILINGUALISM MUL, V323
[10]  
Mansour Saab, P IWSLT 2012