Audio captioning on AudioCaps

18개 결과 · ⬇ CSV · JSON

SPIDEr

Rank	Model	SPIDEr	CIDEr	SPICE	BLEU-4	METEOR	ROUGE-L	FENSE	SPIDEr-FL	Extra Training Data	Paper	Code	Year
1	MQ-Cap	0.519	0.845	0.194	0.301	0.266	–	–	–	✓	Enhancing Retrieval-Augmented Audio Captioning with Generation-Assisted Multimodal Querying and Progressive Learning		2024
2	SLAM-AAC	0.518	0.841	0.194	–	0.268	–	0.668	0.515	✓	SLAM-AAC: Enhancing Audio Captioning with Paraphrasing Augmentation and CLAP-Refine through LLMs	X-LANCE/SLAM-LLM	2024
3	LAVCap	0.517	0.849	0.185	0.297	0.262	0.510	–	–		LAVCap: LLM-based Audio-Visual Captioning using Optimal Transport	naver-intel-co-lab/gaudi-lavcap · Hyeongkeun/LAVCap	2025
4	EnCLAP++-large	0.510	0.823	0.197	–	0.269	–	0.665	–	✓	EnCLAP++: Analyzing the EnCLAP Framework for Optimizing Automated Audio Captioning Performance	jaeyeonkim99/enclap	2024
5	AutoCap	0.507	0.832	0.182	–	0.253	0.518	–	–		Taming Data and Transformers for Audio Generation	snap-research/GenAU	2024
6	LOAE	0.505	0.816	0.193	–	0.267	–	0.664	–	✓	Enhancing Automated Audio Captioning via Large Language Models with Optimized Audio Encoding	frankenliu/LOAE	2024
7	EnCLAP++-base	0.501	0.815	0.188	–	0.257	–	0.661	–	✓	EnCLAP++: Analyzing the EnCLAP Framework for Optimizing Automated Audio Captioning Performance	jaeyeonkim99/enclap	2024
8	EnCLAP-large	0.4954	0.8029	0.1879	–	0.2554	–	–	–		EnCLAP: Combining Neural Audio Codec and Audio-Text Joint Embedding for Automated Audio Captioning	jaeyeonkim99/enclap	2024
9	CNext-trans	0.4951	0.8061	0.1841	–	0.2527	–	0.6431	0.4945
10	EnCLAP-base	0.4829	0.7795	0.1863	–	0.2473	–	–	–		EnCLAP: Combining Neural Audio Codec and Audio-Text Joint Embedding for Automated Audio Captioning	jaeyeonkim99/enclap	2024
11	AL-MixGen + Multi-TTA	0.475	0.769	0.181	–	–	–	–	–
12	Rethink-ACT (AST + TF + MIL)	0.472	0.764	0.180	0.285	0.242	0.504	–	–		Rethinking Transfer and Auxiliary Learning for Improving Audio Captioning Transformer		2023
13	AL-MixGen	0.466	0.755	0.177	–	–	–	–	–		Exploring Train and Test-Time Augmentations for Audio-Language Learning		2022
14	BART + YAMNet + PANNs	0.465	0.753	0.176	–	–	–	–	–		AUTOMATED AUDIO CAPTIONING BY FINE-TUNING BART WITH AUDIOSET TAGS	felixgontier/dcase2021aac	2021
15	CNN+Transformer	0.426	0.693	0.159	–	–	–	–	–		Audio Captioning Transformer	XinhaoMei/ACT	2021
16	TopDown-AlignedAtt (1NN)	0.369	0.593	0.144	–	–	–	–	–		AudioCaps: Generating Captions for Audios in The Wild		2019
17	VAST	–	0.781	–	0.295	0.247	0.509	–	–	✓	VAST: A Vision-Audio-Subtitle-Text Omni-Modality Foundation Model and Dataset	TXH-mercury/VALOR · txh-mercury/vast	2023
18	VALOR	–	0.741	–	0.270	0.231	0.494	–	–	✓	VALOR: Vision-Audio-Language Omni-Perception Pretraining Model and Dataset	TXH-mercury/VALOR	2023