Text-to-Speech on IndicTTS Database Kannada
4.11MOSGround Truth
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Ground Truth2022.11 | 4.11 | — | — | 13.2 | |
| AI4Bharat-TTSAcoustic Model=FastPitch, Vocoder=HiFiGAN2022.11 | 3.68 | 8.47 | 0.27 | 12 | |
| Tacotron2+WaveGlowAcoustic Model=Tacotron2, Vocoder=WaveGlow2022.11 | 2.9 | 13.11 | 0.35 | 21.7 | |
| GlowTTS+HiFiGANAcoustic Model=GlowTTS, Vocoder=HiFiGAN2022.11 | 2.27 | 9.64 | 0.3 | 39.1 |