Abstract | ||
---|---|---|
This paper addresses novel advances in English dialect/accent classification/identification. A word level based modeling technique is proposed that is shown to outperform a LVCSR based system with significantly less computational costs. The new algorithm, which is named WDC (Word based Dialect Classification), converts the text independent decision problem into text dependent problem and produces multiple combination decisions at the word level rather than make a single decision at the utterance level. There are two sets of classifiers employed for WDC: word classifier D-W(k) and utterance classifier D-u. D-W(k) is boosted via real Ada-Boost.MH algorithm in the probability space directly instead of the feature space. D-u is boosted via the dialect dependency information of the words. Two dialect corpora are used in the evaluation. Significant improvement in dialect classification is achieved for both corpora. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1109/ICASSP.2005.1415181 | ICASSP (1) |
Keywords | Field | DocType |
accent classification,accent identification,utterance classifier,speech recognition,natural languages,decision making,word classifier,dialect dependency information,boosted word modeling,text dependent problem,signal classification,word-based dialect classification,dialect identification,feature space,probability space,text independent decision problem,probability,hidden markov models,stress,speech processing,decision problem,robustness | Speech processing,Computer science,Utterance,Natural language processing,Artificial intelligence,Classifier (linguistics),Decision problem,Feature vector,Pattern recognition,Speech recognition,Natural language,Hidden Markov model,Vocabulary | Conference |
Volume | ISSN | ISBN |
1 | 1520-6149 | 0-7803-8874-7 |
Citations | PageRank | References |
6 | 0.51 | 11 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Rongqing Huang | 1 | 141 | 10.27 |
John H. L. Hansen | 2 | 3215 | 365.75 |