Abstract | ||
---|---|---|
Children present a unique challenge to automatic speech recognition. Today's state-of-the-art speech recognition systems still have problems handling children's speech because acoustic models are trained on data collected from adult speech. In this paper we describe an inexpensive way to mend this problem. We collected children's speech when they interact with an automated reading tutor. These data are subsequently transcribed by a speech recognition system and automatically filtered. We studied how to use these automatically collected data to improve children's speech recognition system's performance. Experiments indicate that automatically collected data can reduce the error rate significantly on children's speech. |
Year | Venue | Keywords |
---|---|---|
1998 | ICSLP | speech recognition,data collection,error rate,automatic speech recognition |
Field | DocType | Citations |
Speech corpus,Speech processing,Computer science,Speaker recognition,Natural language processing,Artificial intelligence,Speech analytics,Pattern recognition,Voice activity detection,Audio mining,Word error rate,Speech recognition,Acoustic model | Conference | 3 |
PageRank | References | Authors |
0.66 | 14 | 8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Gregory Aist | 1 | 125 | 29.06 |
Peggy Chan | 2 | 3 | 0.66 |
Xuedong Huang | 3 | 1390 | 283.19 |
Li Jiang | 4 | 3 | 0.66 |
Rebecca Kennedy | 5 | 3 | 0.66 |
DeWitt Latimer IV | 6 | 3 | 0.66 |
Jack Mostow | 7 | 1133 | 263.51 |
Calvin Yeung | 8 | 3 | 0.66 |