List your company

Zero-shot Text-to-Speech on LibriSpeech SNR = 12dB (test-clean)

3.09UTMOSF5-TTS
1.60281.98892.3752.7611May 19, 2025

Evaluation results

MethodLinks
F5-TTSTraining Dataset=LibriTTS2025.053.090.250.29—5042.2460.03
OZSpeechTraining Dataset=LibriTTS, Fine-tuned on noisy prompts=true2025.053.040.050.330.397615730.01
OZSpeechTraining Dataset=LibriTTS2025.052.650.050.280.357021.7530.03
VALL-ETraining Dataset=LibriTTS2025.052.430.510.250.315459.25400.02
VoiceCraftTraining Dataset=GigaSpeech2025.052.420.20.40.45932.48600.01
NaturalSpeech 2Training Dataset=LibriTTS2025.051.660.120.220.347120450.01