Table 3 ASR performance in WERs of conventional MFCC features modeled by hybrid (HMM/DNN)- and BN-HMM/GMM-based systems on MP-FR evaluation data while taking into account different foreign data for training
System | HMM/DNN | BN-HMM/GMM | ||
|---|---|---|---|---|
(training) | Supervised | Semi-supervised | Supervised | Semi-supervised |
| Â | (%) | (%) | (%) | (%) |
MP-FR (19 h) | 14.9 | – | 14.5 | – |
+ESTER-20 (20 h) | 14.3 | 14.8 | 13.9 | 14.0 |
+ESTER-57 (57 h) | 14.0 | 14.5 | 13.7 | 13.9 |
+MP-GE (18 h) | 14.6 | 14.7 | 14.2 | 14.1 |