List your company

Text-to-Speech on ELLA-V hard sentences (test)

4.29WER (%)E1 TTS DMD
4.11845.27676.4357.5933Oct 9, 2024

Evaluation results

MethodLinks
E1 TTS DMDmode=one-step TTS, source=reported in E1 TTS2024.104.291.891.620.74
F5-TTSmode=zero-shot, NFE=32, Sway Sampling=true2024.104.41.812.40.18
StyleTTS 2mode=zero-shot, source=reported in E1 TTS2024.104.832.172.030.61
CosyVoicemode=zero-shot, source=reported in E1 TTS2024.108.33.472.741.93
E2 TTSmode=zero-shot, reproduced=true, NFE=32, Sway Sampling=true2024.108.583.74.820.06