List your company

Automated Audio Captioning on Clotho

55.92AAC ScoreKimi-Audio
-0.052814.478629.0143.5414Sep 24, 2025

Evaluation results

MethodLinks
Kimi-AudioSize=7B2025.0955.92
SLAM-OmniSize=0.5B2025.0954.52
GLM-4-VoiceSize=9B2025.0913.15
TtTSize=3B2025.0912.63
Pretrain+TtTSize=3B2025.0911.55
Qwen2.5-3B (AR)Size=3B2025.099.73
Qwen2.5-3B (NAR)Size=3B2025.099.54
VITA-AudioSize=7B2025.096.18
MoshiSize=7B2025.094.32
Mini-OmniSize=0.5B2025.093.61
LLaMA-OmniSize=8B2025.092.53
SpeechGPTSize=7B2025.092.1