Table 4 Summary and splits of the utilized datasets
From: Accent modification for speech recognition of non-native speakers using neural style transfer
Trained network | Utilized dataset |
|---|---|
Autoencoder | UMEERJ (18,662 samples subset) |
Style transfer network | UMEERJ (split (0.8/0.1/0.1)) |
Loss network | LibriSpeech (train-clean-360) |
CNN-RNN-based ASR | LibriSpeech (train-clean-360) |
TDNN-based ASR | LibriSpeech (train-clean-360) |