AL-MixGen + Multi-TTA

Reported on 5 benchmarks across 2 tasks · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Audio5 results

Text to Audio RetrievalonAudioCaps
R@1· 2022-10-31
34.7
best: 55.2 (InternVideo2-6B)
SOTA
Exploring Train and Test-Time Augmentations for Audio-Language Learning arXiv:2210.17143
Text to Audio RetrievalonAudioCaps
R@10· 2022-10-31
83.3
best: 88.4 (ONE-PEACE)
SOTA
Exploring Train and Test-Time Augmentations for Audio-Language Learning arXiv:2210.17143
Audio captioningonAudioCaps
CIDEr
0.769
best: 50.2 (Audio Flamingo)
Audio captioningonAudioCaps
SPICE
0.181
best: 15.1 (Audio Flamingo)
Audio captioningonAudioCaps
SPIDEr
0.475
best: 32.6 (Audio Flamingo)