{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/slam-aac-enhancing-audio-captioning-with","title":"SLAM-AAC: Enhancing Audio Captioning with Paraphrasing Augmentation and CLAP-Refine through LLMs","arxiv_id":"2410.09503","date":"2024-10-12","proceeding":null,"authors":["Wenxi Chen","Ziyang Ma","Xiquan Li","Xuenan Xu","Yuzhe Liang","Zhisheng Zheng","Kai Yu","Xie Chen"],"abstract":"Automated Audio Captioning (AAC) aims to generate natural textual descriptions for input audio signals. Recent progress in audio pre-trained models and large language models (LLMs) has significantly enhanced audio understanding and textual reasoning capabilities, making improvements in AAC possible. In this paper, we propose SLAM-AAC to further enhance AAC with paraphrasing augmentation and CLAP-Refine through LLMs. Our approach uses the self-supervised EAT model to extract fine-grained audio representations, which are then aligned with textual embeddings via lightweight linear layers. The caption generation LLM is efficiently fine-tuned using the LoRA adapter. Drawing inspiration from the back-translation method in machine translation, we implement paraphrasing augmentation to expand the Clotho dataset during pre-training. This strategy helps alleviate the limitation of scarce audio-text pairs and generates more diverse captions from a small set of audio clips. During inference, we introduce the plug-and-play CLAP-Refine strategy to fully exploit multiple decoding outputs, akin to the n-best rescoring strategy in speech recognition. Using the CLAP model for audio-text similarity calculation, we could select the textual descriptions generated by multiple searching beams that best match the input audio. Experimental results show that SLAM-AAC achieves state-of-the-art performance on Clotho V2 and AudioCaps, surpassing previous mainstream models.","url_abs":"https://arxiv.org/abs/2410.09503v1","url_pdf":"https://arxiv.org/pdf/2410.09503v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"slam-aac-enhancing-audio-captioning-with","repo_url":"https://github.com/X-LANCE/SLAM-LLM","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"audio-captioning","task_name":"Audio captioning"},{"task_slug":null,"task_name":"AudioCaps"},{"task_slug":"caption-generation","task_name":"Caption Generation"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"speech-recognition","task_name":"Speech Recognition"},{"task_slug":"translation","task_name":"Translation"},{"task_slug":"speech-recognition-1","task_name":"speech-recognition"},{"task_slug":"text-similarity","task_name":"text similarity"}],"methods":[{"method_slug":"set","method_name":"SET"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/audio-captioning-on-audiocaps","task":"Audio captioning","dataset":"AudioCaps","model":"SLAM-AAC","rank_in_archive_order":2,"of":18,"metrics":{"CIDEr":"0.841","FENSE":"0.668","METEOR":"0.268","SPICE":"0.194","SPIDEr":"0.518","SPIDEr-FL":"0.515"},"uses_additional_data":true},{"leaderboard":"/sota/audio-captioning-on-clotho","task":"Audio captioning","dataset":"Clotho","model":"SLAM-AAC","rank_in_archive_order":1,"of":11,"metrics":{"CIDEr":"0.515","FENSE":"0.540","METEOR":"0.197","SPICE":"0.148","SPIDEr":"0.332","SPIDEr-FL":"0.330"},"uses_additional_data":true}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2410.09503","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}