List your company

Voice Cloning on SEED-TTS-Eval ZH (test)

0.54CERFish Audio S2
0.4021.33352.2653.1965Dec 15, 2025Jan 12, 2026Feb 10, 2026Mar 11, 2026Apr 8, 2026May 7, 2026Jun 5, 2026

Evaluation results

MethodLinks
Fish Audio S2Parameters=4B, Open-source=true, Zero-shot=true2026.060.54——
MiniMax-SpeechOpen-source=false, Zero-shot=true2026.060.83—78.3
OmniVoiceParameters=0.8B, Open-source=true, Zero-shot=true2026.060.84—77.7
CosyVoice3.5Open-source=false, Zero-shot=true2026.060.87—79.7
VoxCPMParameters=0.6B, Open-source=true, Zero-shot=true2026.060.93—77.2
VoxCPM2Parameters=2B, Open-source=true, Zero-shot=true2026.060.97—79.5
DiTARParameters=0.6B, Open-source=false, Zero-shot=true2026.061.02—75.3
Index-TTS 2Params=1.5B, Category=Multi-Stage or NAR Methods, Independent control of timbre and prosody=true2025.121.030.765—
IndexTTS2Parameters=1.5B, Open-source=true, Zero-shot=true2026.061.03—76.5
Qwen3-OmniParameters=30B-A3B, Open-source=true, Zero-shot=true2026.061.07——
LongCat-Audio-DiTParameters=3.5B, Open-source=true, Zero-shot=true2026.061.09—81.8
CosyVoice 3Parameters=1.5B, Open-source=false, Zero-shot=true2026.061.12—78.1
Seed-TTSOpen-source=false, Zero-shot=true2026.061.12—79.6
FireRedTTS-2Parameters=1.5B, Open-source=true, Zero-shot=true2026.061.14—73.6
CosyVoice 3Parameters=0.5B, Open-source=true, Zero-shot=true2026.061.16—78
VibeVoiceParameters=1.5B, Open-source=true, Zero-shot=true2026.061.16—74.4
OpenAudio-s1-miniParameters=0.5B, Open-source=true, Zero-shot=true2026.061.18—68.5
VoxCPM1.5Parameters=0.8B, Open-source=true, Zero-shot=true2026.061.18—77
Spark-TTSParams=0.5B, Category=One-Stage AR Methods, Independent control of timbre and prosody=false2025.121.20.672—
MOSS-TTSParameters=8B, Open-source=true, Zero-shot=true2026.061.2—78.8
Qwen3-TTSParameters=1.7B, Open-source=true, Zero-shot=true2026.061.22—77
CosyVoice 2Parameters=0.5B, Open-source=true, Zero-shot=true2026.061.38—75.7
ZipVoiceParameters=0.1B, Open-source=true, Zero-shot=true2026.061.4—75.1
CosyVoice2Params=0.5B, Category=Multi-Stage or NAR Methods, Independent control of timbre and prosody=false2025.121.450.748—
HiggsAudio-v2Parameters=3B, Open-source=true, Zero-shot=true2026.061.5—74
FireRedTTSParams=N/A, Category=Multi-Stage or NAR Methods, Independent control of timbre and prosody=false2025.121.510.635—
FireRedTTSParameters=0.5B, Open-source=true, Zero-shot=true2026.061.51—63.5
MegaTTS3Parameters=0.5B, Open-source=false, Zero-shot=true2026.061.52—79
F5-TTSParameters=0.3B, Open-source=true, Zero-shot=true2026.061.53—76
Spark-TTSParameters=0.5B, Open-source=true, Zero-shot=true2026.061.54—66
F5-TTSParams=0.3B, Category=Multi-Stage or NAR Methods, Independent control of timbre and prosody=false2025.121.560.741—
Qwen2.5-OmniParameters=7B, Open-source=true, Zero-shot=true2026.061.7—75.2
Llasa-1BParams=1B, Category=One-Stage AR Methods, Independent control of timbre and prosody=false2025.121.890.669—
DisCo-SpeechParams=1.5B, Category=One-Stage AR Methods, Independent control of timbre and prosody=true2025.122.080.677—
MaskGCTParameters=1B, Open-source=true, Zero-shot=true2026.062.27—77.4
CosyVoiceParameters=0.3B, Open-source=true, Zero-shot=true2026.063.63—72.3
VevoParams=N/A, Category=Multi-Stage or NAR Methods, Independent control of timbre and prosody=true2025.123.990.723—