List your company

Speech Synthesis on Minimax Multilingual (test)

0.73Chinese WER (%)Fish Audio S2
0.1184.2498.3812.511Mar 9, 2026Mar 23, 2026Apr 7, 2026Apr 22, 2026May 6, 2026May 21, 2026Jun 5, 2026

Evaluation results

MethodLinks
Fish Audio S22026.030.733.50.7530.670.8050.8162.840.7980.990.731.620.7973.330.8193.050.6980.550.7675.740.79514.640.8211.460.7631.270.7472.760.7961.180.8171.260.8191.140.78110.740.7332.40.790.910.7764.230.7860.870.8352.30.7477.410.74
Fish Audio S2ASR evaluation model=Whisper-large-v32026.060.733.5—30.67——2.84—0.99—1.62—3.33—3.05—0.55—5.74—14.64—1.46—1.27—2.76—1.18—1.26—1.14—10.74—2.4—0.91—4.23—0.87—2.3—7.41—
Qwen3-TTSASR evaluation model=Whisper-large-v32026.060.93—————————0.93———2.86—1.24———————0.95—3.82—1.76———1.53———3.21—1.13—————————
Fish Audio S12026.030.986.420.71348.80.7730.77718.020.7332.270.7012.370.785100.785.70.6020.650.71919.840.74632.280.79280.6961.870.7143.450.781.990.7473.410.7941.940.75619.490.7395.250.7641.780.75396.750.6912.260.78614.490.65378.130.664
VoxCPM2ASR evaluation model=Whisper-large-v32026.061.1413.05—38.58——24.13—0.91—2.29—2.63—4.53—0.68—2.84—19.7—1.08—1.56—4.63—1.96—1.14—1.94—21.58—3.63—1.44—2.96—0.82—6.32—3.31—
MinimaxASR evaluation model=Whisper-large-v32026.062.251.67—34.11——3.88—1.14—2.16—4.67—4.1—1.91—2.02—6.96—1.24—1.54—3.52—1.75—1.42—1.88—2.88—4.28—1.03—2.7—1.52—1.08—0.88—
MiniMax2026.032.2521.6650.73634.1110.7780.783.8750.7961.1430.7382.1640.7564.6660.8354.0990.6281.9060.7332.0160.8266.9620.8181.2370.7291.5430.6993.5190.7761.7470.7761.4150.8021.8770.8052.8780.8094.2810.7611.0290.7622.7010.81.520.7791.0820.730.880.743
ElevenLabs2026.0316.0261.6660.70651.5130.670.6772.1080.6850.8030.682.3390.6132.9640.7595.2160.5350.5720.6140.9910.7335.8270.731.0590.661.7430.57910.6460.7381.8650.70.7660.7291.3310.7111.3470.6993.8780.6761.0840.61573.9360.5880.6990.5960.9970.64773.4150.369
ElevenLabsASR evaluation model=Whisper-large-v32026.0616.031.67—51.51——2.11—0.8—2.34—2.96—5.22—0.57—0.99—5.83—1.06—1.74—10.65—1.87—0.77—1.33—1.35—3.88—1.08—73.94—0.7—1—73.42—