List your company

Text-to-speech on Seed-TTS (eval)

1.85WERVoxCPM
1.35.01258.72512.4375Dec 29, 2025Jan 6, 2026Jan 15, 2026Jan 23, 2026Feb 1, 2026Feb 9, 2026Feb 18, 2026

Evaluation results

MethodLinks
VoxCPMBitrate Control=false, Architecture=NAR / Continuous AR2026.021.8572.9
CAT-TTSBitrate Control=true, Architecture=Discrete AR2026.021.8973.1
Ground Truth2026.011.9—
GLM-TTSBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.021.9168.1
OpenAudio-s1-miniBitrate Control=false, Architecture=Discrete AR2026.021.9455
FireRedTTS2Bitrate Control=false, Architecture=Discrete AR2026.021.9566.5
SparkTTSBitrate Control=false, Architecture=Discrete AR2026.021.9858.4
F5-TTSBitrate Control=false, Architecture=NAR / Continuous AR2026.02267
CosyVoice 22026.012.10.66
CosyVoice3-1.5BBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.022.2272
IndexTTS2Bitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.022.2370.6
HiggsAudio-v2Bitrate Control=false, Architecture=Discrete AR2026.022.4467.7
MaskGCTBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.022.6271.7
FlexiCodec-TTSBitrate Control=true, Architecture=Cascade (AR+NAR / NAR+NAR)2026.022.6365.7
Qwen2.5-OmniBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.022.7263.2
LlasaBitrate Control=false, Architecture=Discrete AR2026.022.9757.4
VibeVoiceBitrate Control=false, Architecture=NAR / Continuous AR2026.023.0468.9
CosyVoice2Bitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.023.0965.9
Step-Audio2-mini2025.123.18—
FireRedTTSBitrate Control=false, Architecture=Cascade (AR+NAR / NAR+NAR)2026.023.8446
KanadeToken Rate=25Hz2026.0140.48
KMToken Rate=25Hz2026.014.90.42
MiMo-Audio-7B-Instruct2025.125.37—
KMToken Rate=12.5Hz2026.015.40.42
DualCodec2026.015.50.34
KanadeToken Rate=12.5Hz2026.015.70.47
Mimi2026.0160.32
SODA-base-4BScale=4B, Shot count=0-shot2026.026.10.56
SODA-base-1.7BScale=1.7B, Shot count=0-shot2026.026.90.56
X-Codec 22026.017.20.35
BiCodec2026.017.50.46
SODA-base-600MScale=600M, Shot count=0-shot2026.027.60.555
PAST2026.0190.35
SODA-prelim-600MScale=600M, Shot count=0-shot2026.029.20.516
StableCodec2026.0110.90.23
ST2026.0111.20.35
SODA-base-135MScale=135M, Shot count=0-shot2026.0211.20.5
TiCodec2026.0112.90.33
WavTokenizer2026.0115.60.28