List your company

Complex Instruction Following on InstructTTSEval EN

96.2APSGround-truth
53.24864.39975.5586.701Jan 8, 2026Feb 1, 2026Feb 26, 2026Mar 23, 2026Apr 16, 2026May 11, 2026Jun 5, 2026

Evaluation results

MethodLinks
Ground-truthSource Type=Baseline2026.0196.289.467.284.3
Gemini-flashSource Type=Closed-sourced2026.0192.393.880.188.7
Gemini-proSource Type=Closed-sourced2026.0187.68667.280.3
Gemini-TTS-ProVoice Control Type=fixed voice editing2026.0387.68667.2—
Gemini-TTS-Pro2026.0687.68667.2—
VoxCPM22026.0684.283.271.4—
HumeSource Type=Closed-sourced2026.018375.354.371.1
Hume2026.068375.354.3—
Qwen3-TTS-12Hz-1.7B-VDFrequency=12Hz, Parameters=1.7B2026.0682.982.468.4—
FlexiVoiceSource Type=Open-sourced2026.0181.285.271.479.3
MiMo-Audio-7B-InstructSource Type=Open-sourced2026.0180.677.659.572.6
MIMO-Audio-7B-Instruct2026.0380.677.659.5—
Mimo-Audio-7B-InstructParameters=7B2026.0680.677.659.5—
Qwen3-TTS-VD2026.0378.478.872—
GPT-4o-mini-TTSSource Type=Closed-sourced2026.0176.474.354.868.5
GPT-4o-mini-TTSVoice Control Type=fixed voice editing2026.0376.474.354.8—
GPT-4o-mini-TTS2026.0676.474.354.8—
MOSS-VoiceGenerator2026.0368.28268.7—
MOSS-VoiceGenerator2026.0668.28268.7—
PromptTTSSource Type=Open-sourced2026.0164.347.231.447.6
PromptTTS2026.0664.347.231.4—
FlexiVoice-BaseSource Type=Open-sourced2026.0163.67560.666.4
ParlerTTSSource Type=Open-sourced2026.016045.931.245.7
Parler-TTS-largeScale=large2026.066045.931.2—
PromptStyleSource Type=Open-sourced2026.0157.446.430.938.2
PromptStyle2026.0657.446.430.9—
VoxInstructSource Type=Open-sourced2026.0154.95739.350.4
VoxInstructRequires reference audio=true2026.0354.95739.3—
VoxInstruct2026.0654.95739.3—