Audio Captioning on AudioCaps AudioSet (test)
48.5SPIDErHTSAT-BART
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| HTSAT-BART2025.03 | 48.5 | 25 | 78.7 | 18.2 | — | |
| CNN14-BART2025.03 | 46.8 | 24.7 | 75.6 | 17.9 | — | |
| CLAP2023 (Reference)2025.03 | 45.5 | — | — | — | — | |
| M2D-CLAP2025Audio encoder AudioSet fine-tuning=true2025.03 | 45 | 24.3 | 72.4 | 17.6 | 63.9 | |
| Cacophony2025.03 | 44.8 | 23.6 | 72.8 | 16.8 | — | |
| LAION-CLAPAudio encoder AudioSet fine-tuning=true2025.03 | 44.6 | 23.8 | 71.7 | 17.6 | 64 | |
| M2D-CLAP2025 stage2Audio encoder AudioSet fine-tuning=false2025.03 | 43.4 | 23.7 | 69.6 | 17.2 | 62.5 | |
| M2D-CLAP2025 stage2.1Audio encoder AudioSet fine-tuning=false2025.03 | 43.2 | 23.6 | 69.4 | 17.1 | 62.3 | |
| M2D-CLAP2025 stage1Audio encoder AudioSet fine-tuning=false2025.03 | 42.6 | 23.3 | 68.3 | 16.9 | 61.9 | |
| CLAP2023Audio encoder AudioSet fine-tuning=false2025.03 | 40.7 | 22.4 | 65.2 | 16.3 | 60.5 |