ICASSP 2015accepted0 citations

Low-resource keyword search strategies for tamil

Nancy F. Chen, Chongjia Ni, I-Fan Chen, Sunil Sivadas, Van Tung Pham, Haihua Xu, Xiong Xiao, Tze Siong Lau

Abstract

We propose strategies for a state-of-the-art keyword search (KWS) system developed by the SINGA team in the context of the 2014 NIST Open Keyword Search Evaluation (OpenKWS14) using conversational Tamil provided by the IARPA Babel program. To tackle low-resource challenges and the rich morphological nature of Tamil, we present highlights of our current KWS system, including: (1) Submodular optimization data selection to maximize acoustic diversity through Gaussian component indexed N-grams; (2) Keywordaware language modeling; (3) Subword modeling of morphemes and homophones.

BibTeX
@inproceedings{icassp2015_lowresourcekeywo,
  title = {Low-resource keyword search strategies for tamil},
  author = {Nancy F. Chen and Chongjia Ni and I-Fan Chen and Sunil Sivadas and Van Tung Pham and Haihua Xu and Xiong Xiao and Tze Siong Lau and Su Jun Leow and Boon Pang Lim and Cheung-Chi Leung and Lei Wang and Chin-Hui Lee and Alvina Goh and Engsiong Chng and Bin Ma and Haizhou Li},
  booktitle = {ICASSP 2015},
  year = {2015}
}
Low-resource keyword search strategies for tamil · ICASSP 2015