Abstract | ||
---|---|---|
We propose strategies for a state-of-the-art Vietnamese keyword search (KWS) system developed at the Institute for Infocomm Research (I2R). The KWS system exploits acoustic features characterizing creaky voice quality peculiar to lexical tones in Vietnamese, a minimal-resource transliteration framework to alleviate out-of-vocabulary issues from foreign loan words, and a proposed system combination scheme FusionX. We show that the proposed creaky voice quality features complement pitch-related features, reaching fusion gains of 17.7% relative (6.9% absolute). To the best of our knowledge, the proposed transliteration framework is the first reported rule-based system for Vietnamese; it outperforms statistical-approach baselines up to 14.93-36.73% relative on foreign loan word search tasks. Using FusionX to combine 3 sub-systems, the actual term-weighted value (ATWV) reaches 0.4742, exceeding the ATWV=0.3 benchmark for IARPA Babel participants in the NIST OpenKWSB Evaluation. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/ICASSP.2014.6854377 | Acoustics, Speech and Signal Processing |
Keywords | DocType | ISSN |
information retrieval,knowledge based systems,natural language processing,sensor fusion,speech recognition,ATWV,FusionX scheme,Institute for Infocomm Research,KWS system,NIST OpenKWSB Evaluation,Vietnamese keyword search,acoustic features,actual term-weighted value,creaky voice quality feature,fusion gain,lexical tones,minimal-resource transliteration framework,pitch-related feature,rule-based system,audio indexing,deep neural networks (DNN),glottalization,large vocabulary continuous speech recognition (LVCSR),low-resourced languages,spoken term detection | Conference | 1520-6149 |
Citations | PageRank | References |
2 | 0.41 | 0 |
Authors | ||
8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Nancy F. Chen | 1 | 120 | 28.98 |
Sunil Sivadas | 2 | 31 | 2.19 |
Boon Pang Lim | 3 | 62 | 8.89 |
Hoang Gia Ngo | 4 | 8 | 4.21 |
Haihua Xu | 5 | 55 | 11.41 |
Van Tung Pham | 6 | 40 | 8.42 |
Bin Ma | 7 | 44 | 4.45 |
Haizhou Li | 8 | 3678 | 334.61 |