LSTM-RNN parametric

Reported on 6 benchmarks across 3 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio6 results

Speech RecognitiononMandarin Chinese
Mean Opinion Score· 2016-09-12
3.79
best: 4.08 (WaveNet (L+F))
WaveNet: A Generative Model for Raw Audio arXiv:1609.03499
Speech RecognitiononNorth American English
Mean Opinion Score· 2016-09-12
3.67
best: 4.526 (Tacotron 2)
WaveNet: A Generative Model for Raw Audio arXiv:1609.03499
Speech SynthesisonMandarin Chinese
Mean Opinion Score· 2016-09-12
3.79
best: 4.08 (WaveNet (L+F))
WaveNet: A Generative Model for Raw Audio arXiv:1609.03499
Speech SynthesisonNorth American English
Mean Opinion Score· 2016-09-12
3.67
best: 4.526 (Tacotron 2)
WaveNet: A Generative Model for Raw Audio arXiv:1609.03499
Accented Speech RecognitiononMandarin Chinese
Mean Opinion Score· 2016-09-12
3.79
best: 4.08 (WaveNet (L+F))
WaveNet: A Generative Model for Raw Audio arXiv:1609.03499
Accented Speech RecognitiononNorth American English
Mean Opinion Score· 2016-09-12
3.67
best: 4.526 (Tacotron 2)
WaveNet: A Generative Model for Raw Audio arXiv:1609.03499