List your company

Text-to-Speech on Seed-TTS en (test)

0.8WERFish Speech
0.7421.13351.5251.9165Oct 9, 2024Jan 23, 2025May 9, 2025Aug 23, 2025Dec 7, 2025Mar 23, 2026Jul 7, 2026

Evaluation results

MethodLinks
Fish SpeechType=TTS2026.030.8————————————————
WAND (on IndexTTS 1.5)Attn. Mechanism=WAND (W = 32), Cache=O(1), Compute=O(1)2026.030.91————4.133.980.944—13.0166.23.281.89————
IndexTTS 1.5Attn. Mechanism=MHA (W = ∞), Cache=O(L), Compute=O(L)2026.030.98————4.133.970.944—38.44—6.18—————
Qwen3-TTS-12Hz-1.7B-Basezero-shot=true, frame_rate=12Hz, parameters=1.7B2026.011.24————————————————
F5-TTS#Data(h)=100k Multi., #Params=336M2026.031.26—————3.69—0.67————————
Qwen3.5-Omni-PlusSetting=Zero-shot, Aspect=Content Consistency2026.041.26————————————————
Qwen3-TTS-12Hz-0.6B-Basezero-shot=true, frame_rate=12Hz, parameters=0.6B2026.011.32————————————————
VoXtream2#Data(h)=40k EN, #Params=462M2026.031.32—————4.05—0.656————————
Kyutai-TTS#Data(h)=88k EN, #Params=750M2026.031.34—————3.63—0.689————————
Qwen3-Omni2026.021.39————————————————
Qwen3-Omni-30B-A3BSetting=Zero-shot, Aspect=Content Consistency2026.041.39————————————————
VoxCPM2 + SR-FDSteps=42026.071.410.76————3.76————————3.073.65
GroundtruthAttn. Mechanism=−, Cache=−, Compute=−2026.031.44————3.793.531—————————
CosyVoice 32026.021.45————————————————
CosyVoice 3zero-shot=true2026.011.45————————————————
VoXtream2-B#Data(h)=40k EN, #Params=462M2026.031.45—————3.56—0.679————————
CosyVoice 3Setting=Zero-shot, Aspect=Content Consistency2026.041.45————————————————
GroundtruthHybrid=–, Source=–, Params=–, Layers=–, Data (kh)=–, Ratio=–, Strategy=–, TFLOPs=–2026.031.47————3.933.521—————————
ARCHI-TTSSteps=322026.071.47————————————————
Qwen3-TTS-25Hz-1.7B-Basezero-shot=true, frame_rate=25Hz, parameters=1.7B2026.011.49————————————————
Human#Data(h)=-, #Params=-2026.031.5—————3.53—0.734————————
WavTTSParams=673M, Data (hrs)=100K Emilia, Protocol=Zero-shot2026.061.50.65————3.92——————————
ERNIE 5.02026.021.54————————————————
Qwen3-TTSParams.=1.1B, Training Data (hours)=5000k Multi.2026.041.540.708————4.16——————————
Qwen3-TTSZero-shot=true, #Params=1.1B2026.051.540.708————4.16——————————
OmniVoiceParams.=0.8B, Training Data (hours)=581k Multi.2026.041.60.741————3.91——————————
OmniVoiceZero-shot=true, #Params=0.8B2026.051.60.741————3.91——————————
ZipVoiceParams=123M, Data (hrs)=100K Emilia, Protocol=Zero-shot2026.061.60.7————3.83——————————
Qwen3-TTS-25Hz-0.6B-Basezero-shot=true, frame_rate=25Hz, parameters=0.6B2026.011.64————————————————
MiniMax-Speechzero-shot=true2026.011.65————————————————
MiniMax-SpeechSetting=Zero-shot, Aspect=Content Consistency2026.041.65————————————————
F5-TTSParams=336M, Data (hrs)=100K Emilia, Protocol=Zero-shot2026.061.650.66————3.73——————————
ZipVoiceParams.=0.1B, Training Data (hours)=100k Emilia2026.041.70.697————3.82——————————
ZipVoiceZero-shot=true, #Params=0.1B2026.051.70.697————3.82——————————
OmniVoice-EmiliaParams.=0.8B, Training Data (hours)=100k Emilia2026.041.720.717————3.88——————————
WAND (on CosyVoice2-0.5B)Attn. Mechanism=WAND (W = 32), Cache=O(1), Compute=O(1)2026.031.72————4.024.210.945—5.2549.97.441.55————
OmniVoice-EmiliaZero-shot=true, #Params=0.8B2026.051.720.717————3.88——————————
Seed-TTS DiT2024.101.7330.79———————————————
VoxCPM2Steps=102026.071.740.76————3.81————————3.093.67
DELTA-TTSType=Ours, Params.=0.5B+94M, Training Data (hours)=0.585K LibriTTS2026.071.750.688———————————————
Ground TruthParams=–, Data (hrs)=–, Protocol=Zero-shot2026.061.790.73————3.53——————————
F5-TTSType=TTS2026.031.8————————————————
F5-TTSNFE=322024.101.830.670.313.89—————————————
F5-TTS2026.021.83————————————————
F5-TTSzero-shot=true2026.011.83————————————————
F5-TTSSetting=Zero-shot, Aspect=Content Consistency2026.041.83————————————————
F5-TTS2026.061.830.647———————————————
F5-TTSSteps=322026.071.83————————————————
Spark-TTS#Data(h)=102k Multi., #Params=507M2026.031.85—————3.94—0.572————————
F5-TTSParams.=0.4B, Training Data (hours)=100K Emilia2026.041.850.664————3.72——————————
F5-TTSZero-shot=true, #Params=0.4B2026.051.850.664————3.72——————————
VoxCPMType=AR, Params.=0.5B, Training Data (hours)=1800K Multi.2026.071.850.729———————————————
F5-TTSNFE=162024.101.890.670.163.79—————————————
VoxCPMParams.=0.7B, Training Data (hours)=1800k Multi.2026.041.920.731————3.77——————————
VoxCPMZero-shot=true, #Params=0.7B2026.051.920.731————3.77——————————
Seed-TTS_RL2026.021.94————————————————
KALL-Ezero-shot=true2026.011.94————————————————
CosyVoice2-0.5BAttn. Mechanism=GQA (W = ∞), Cache=O(L), Compute=O(L)2026.031.94————4.014.180.946—10.48—11.55—————
Seed-TTS RLSetting=Zero-shot, Aspect=Content Consistency2026.041.94————————————————
LongCat-AudioDiTParams=1420M, Data (hrs)=100K Multi., Protocol=Zero-shot2026.061.940.76————3.8——————————
FireRedTTS 2zero-shot=true2026.011.95————————————————
FireRedTTS2Type=AR, Params.=1.5B, Training Data (hours)=1400K Multi.2026.071.950.665———————————————
Chatterbox-FlashZero-shot=true, Decoding method=PMI (α = 0), #Params=0.5B, Steps=82026.051.960.704————4.09——————————
Chatterbox-FlashZero-shot=true, Decoding method=TS schedule, #Params=0.5B, Steps=82026.051.970.703————4.09——————————
Spark TTS2026.021.98————————————————
Spark TTSAccess=Open-Source2026.021.98————————————————
Spark TTSzero-shot=true2026.011.98————————————————
Spark TTSSetting=Zero-shot, Aspect=Content Consistency2026.041.98————————————————
Spark-TTSParams=507M, Data (hrs)=102K Multi., Protocol=Zero-shot2026.061.980.58———————————————
Spark-TTS2026.061.980.584———————————————
Spark TTSType=AR, Params.=0.5B, Training Data (hours)=100K Multi.2026.071.980.573———————————————
F5-TTSType=NAR, Params.=0.3B, Training Data (hours)=100K Emilia, NFE=322026.0720.647———————————————
CosyVoice3Type=AR, Params.=0.5B, Training Data (hours)=1000K Multi.2026.072.020.692———————————————
CosyVoice 2Hybrid=×, Source=Public, Params=0.5B, Layers=24, Data (kh)=170, Ratio=1:0, Strategy=–, TFLOPs=1.782026.032.03————3.684.150.66—————————
MaskGCT#Data(h)=100k Multi., #Params=1048M2026.032.03—————3.56—0.712————————
Chatterbox-FlashZero-shot=true, Decoding method=PMI + ED (α = 0.5), #Params=0.5B, Steps=6.42026.052.040.704————4.08——————————
Ground Truth2024.102.060.7303.91—————————————
Vocoder Resynthesized2024.102.090.7———————————————
Human2026.022.14————————————————
Ground-truth2026.042.140.734————3.52——————————
Ground-truthZero-shot=true2026.052.140.734————3.52——————————
Ground-truthType=–2026.072.140.734———————————————
CosyVoice3Params.=1.1B, Training Data (hours)=1000k Multi.2026.042.170.696————3.96——————————
CosyVoice3Zero-shot=true, #Params=1.1B2026.052.170.696————3.96——————————
E2 TTSNFE=322024.102.190.710.063.81—————————————
E2 TTS2026.022.19————————————————
E2 TTSzero-shot=true2026.012.19————————————————
E2 TTSSetting=Zero-shot, Aspect=Content Consistency2026.042.19————————————————
E2 TTSType=NAR, Params.=0.3B, Training Data (hours)=100K Emilia, NFE=322026.072.190.71———————————————
VoiceStar#Data(h)=65k EN, #Params=840M2026.032.2—————3.92—0.605————————
ChatterboxZero-shot=true, #Params=0.5B2026.052.20.685————4.1——————————
E2-TTSParams=333M, Data (hrs)=100K Emilia, Protocol=Zero-shot2026.062.210.71————3.2——————————
FlashTTSStage=12026.062.210.572———————————————
IndexTTS2Type=AR, Params.=1.5B, Training Data (hours)=55K Emilia2026.072.230.706———————————————
VoxCPM2Steps=42026.072.230.74————3.3————————2.93.53
Seed-TTS_ICL2026.022.24————————————————
Seed-TTS ICLSetting=Zero-shot, Aspect=Content Consistency2026.042.24————————————————
Seed-TTSAccess=Proprietary2026.022.25————————————————
Seed-TTSzero-shot=true2026.012.25————————————————
Seed-TTS2026.062.250.762———————————————