Name
Abstract | ||
|---|---|---|
In this paper, we investigate the problem of automatically selecting textual keywords for keyword search development and tuning on audio data for any language. Briefly, the method samples candidate keywords in the training data while trying to match a set of target marginal distributions for keyword features such as keyword frequency in the training or development audio, keyword length, frequency of out-of-vocabulary words, and TF-IDF scores. The method is evaluated on four IARPA Babel program base period languages. We show the use of the automatically selected keywords for the keyword search system development and tuning. We show also that search performance is improved by tuning the decision threshold on the automatically selected keywords. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1109/ICASSP.2014.6855126 | ICASSP |
Keywords | Field | DocType |
keyword search,speech processing,keyword search turning,automatic textual keyword selection,tf-idf scores,speech recognition,vocabulary,target marginal distributions,spoken term detection,iarpa babel program base period languages,keyword length,audio data,development audio,natural language processing,audio signal processing,query selection,keyword search development,keyword frequency,keyword features,keyword selection,out-of-vocabulary word frequency,query processing,training data,training audio,acoustics,tuning,nist,speech | Keyword density,Training set,Information retrieval,Computer science,Keyword search,Natural language processing,Artificial intelligence,System development,Marginal distribution | Conference |
ISSN | Citations | PageRank |
1520-6149 | 7 | 0.43 |
References | Authors | |
8 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Jia Cui | 1 | 94 | 6.26 |
| Jonathan Mamou | 2 | 527 | 28.72 |
| B. Kingsbury | 3 | 4175 | 335.43 |
| Bhuvana Ramabhadran | 4 | 1779 | 153.83 |