List your company

Zero-shot Text-to-Speech on LibriSpeech SNR = 6dB (test-clean)

2.9UTMOSOZSpeech
1.36081.76042.162.5596May 19, 2025

Evaluation results

MethodLinks
OZSpeechTraining Dataset=LibriTTS, Fine-tuned on noisy prompts=true2025.052.960.290.347217.41740.01
F5-TTSTraining Dataset=LibriTTS2025.052.41270.29—5540.95500.11
OZSpeechTraining Dataset=LibriTTS2025.052.2160.230.296132.8460.05
VoiceCraftTraining Dataset=GigaSpeech2025.051.8270.330.365045.31680.01
VALL-ETraining Dataset=LibriTTS2025.051.66770.140.184096.93440.02
NaturalSpeech 2Training Dataset=LibriTTS2025.051.42160.170.36127.41580.01