Text-to-Speech on Seed-TTS 24 kHz (test-zh)
0.762SIM-oM3-TTS-Fbank
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| M3-TTS-FbankNFE=32, Data=100K Emilia, Params=355M2025.12 | 0.762 | 1.36 | 3.18 | |
| Ground Truth2025.12 | 0.755 | 1.25 | 2.78 | |
| ZipVoiceNFE=16, Data=100K Emilia, Params=123M2025.12 | 0.751 | 1.4 | 3.15 | |
| F5-TTSNFE=32, Data=100K Emilia, Params=336M2025.12 | 0.75 | 1.53 | 2.93 | |
| CosyVoice 2Data=167K Multi., Params=618M2025.12 | 0.748 | 1.45 | — | |
| F5-TTSNFE=32, Data=100K Emilia, Params=155M2025.12 | 0.733 | 1.57 | 2.93 | |
| CosyVoiceData=170K Multi., Params=416M2025.12 | 0.723 | 3.63 | — | |
| E2-TTSNFE=32, Data=100K Emilia, Params=333M2025.12 | 0.713 | 1.91 | 2.26 | |
| VAE Reconstruction2025.12 | 0.699 | 1.35 | 1.89 | |
| Spark-TTSData=102K Multi., Params=507M2025.12 | 0.672 | 1.2 | — | |
| M3-TTS-VAENFE=32, Data=100K Emilia, Params=355M2025.12 | 0.621 | 1.31 | 2.18 |