SLAM-AAC: Enhancing Audio Captioning with Paraphrasing Augmentation and CLAP-Refine through LLMs
Open paper
Benchmark
Model
Rank
Results
audio-captioning-on-audiocaps
SLAM-AAC
#1
SPIDEr: 0.518
CIDEr: 0.841
SPICE: 0.194
METEOR: 0.268
FENSE: 0.668
…
audio-captioning-on-clotho
SLAM-AAC
#1
SPIDEr: 0.332
CIDEr: 0.515
SPICE: 0.148
METEOR: 0.197
FENSE: 0.540
…
Rank counts only results with a code link.