Abstract | ||
---|---|---|
Automatic Speech Recognition (ASR) in reverberant rooms can be improved by choosing training data from the same acoustical environment as the test data. In a real-world application this is often not possible. A solution for this problem is to use speech signals from a close-talking microphone and reverberate them artificially with multiple room impulse responses. This paper shows results on recognizers whose training data differ in size and percentage of reverberated signals in order to find the best combination for data sets with different degrees of reverberation. The average error rate on a close-talking and a distant-talking test set could thus be reduced by 29% relative. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1007/11551874_29 | TSD |
Keywords | Field | DocType |
acoustical environment,training data,distant-talking test set,different degree,test data,automatic speech recognition,close-talking microphone,average error rate,distant-talking asr,best combination,error rate | Speech processing,Reverberation,Computer science,Word error rate,Microphone array,Speech recognition,Test data,Room acoustics,Microphone,Test set | Conference |
Volume | ISSN | ISBN |
3658 | 0302-9743 | 3-540-28789-2 |
Citations | PageRank | References |
7 | 0.82 | 5 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Tino Haderlein | 1 | 85 | 17.32 |
Elmar Nöth | 2 | 959 | 158.94 |
Wolfgang Herbordt | 3 | 45 | 5.77 |
Walter Kellermann | 4 | 535 | 45.32 |
Heinrich Niemann | 5 | 1650 | 288.56 |