Discriminative Approach to Build Hybrid Vocabulary for Conversational Telephone Speech Recognition of Agglutinative Languages

Xin LI  Jielin PAN  Qingwei ZHAO  Yonghong YAN  

IEICE TRANSACTIONS on Information and Systems   Vol.E96-D   No.11   pp.2478-2482
Publication Date: 2013/11/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.E96.D.2478
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Speech and Hearing
agglutinative languages,  speech recognition,  sub-words,  discriminative learning,  hybrid system,  

Full Text: PDF(554.1KB)
>>Buy this Article

Morphemes, which are obtained from morphological parsing, and statistical sub-words, which are derived from data-driven splitting, are commonly used as the recognition units for speech recognition of agglutinative languages. In this letter, we propose a discriminative approach to select the splitting result, which is more likely to improve the recognizer's performance, for each distinct word type. An objective function which involves the unigram language model (LM) probability and the count of misrecognized phones on the acoustic training data is defined and minimized. After determining the splitting result for each word in the text corpus, we select the frequent units to build a hybrid vocabulary including morphemes and statistical sub-words. Compared to a statistical sub-word based system, the hybrid system achieves 0.8% letter error rates (LERs) reduction on the test set.