video-SALMONN: Speech-Enhanced Audio-Visual Large Language Models

Guangzhi Sun, Wenyi Yu, Changli Tang, Xianzhao Chen, Tian Tan, Wei Li, Lu Lu, Zejun Ma, Yuxuan Wang, Chao Zhang
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:47198-47217, 2024.

Abstract

Speech understanding as an element of the more generic video understanding using audio-visual large language models (av-LLMs) is a crucial yet understudied aspect. This paper proposes video-SALMONN, a single end-to-end av-LLM for video processing, which can understand not only visual frame sequences, audio events and music, but speech as well. To obtain fine-grained temporal information required by speech understanding, while keeping efficient for other video elements, this paper proposes a novel multi-resolution causal Q-Former (MRC Q-Former) structure to connect pre-trained audio-visual encoders and the backbone large language model. Moreover, dedicated training approaches including the diversity loss and the unpaired audio-visual mixed training scheme are proposed to avoid frames or modality dominance. On the introduced audio-visual evaluation benchmark, video-SALMONN achieves more than 25% absolute accuracy improvements on the video-QA task and over 30% absolute accuracy improvements on audio-visual QA tasks with human speech. In addition, video-SALMONN demonstrates remarkable video comprehension and reasoning abilities on tasks that are unprecedented by other av-LLMs. Our training code and model checkpoints are available at https://github.com/bytedance/SALMONN/

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-sun24l, title = {video-{SALMONN}: Speech-Enhanced Audio-Visual Large Language Models}, author = {Sun, Guangzhi and Yu, Wenyi and Tang, Changli and Chen, Xianzhao and Tan, Tian and Li, Wei and Lu, Lu and Ma, Zejun and Wang, Yuxuan and Zhang, Chao}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {47198--47217}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/sun24l/sun24l.pdf}, url = {https://proceedings.mlr.press/v235/sun24l.html}, abstract = {Speech understanding as an element of the more generic video understanding using audio-visual large language models (av-LLMs) is a crucial yet understudied aspect. This paper proposes video-SALMONN, a single end-to-end av-LLM for video processing, which can understand not only visual frame sequences, audio events and music, but speech as well. To obtain fine-grained temporal information required by speech understanding, while keeping efficient for other video elements, this paper proposes a novel multi-resolution causal Q-Former (MRC Q-Former) structure to connect pre-trained audio-visual encoders and the backbone large language model. Moreover, dedicated training approaches including the diversity loss and the unpaired audio-visual mixed training scheme are proposed to avoid frames or modality dominance. On the introduced audio-visual evaluation benchmark, video-SALMONN achieves more than 25% absolute accuracy improvements on the video-QA task and over 30% absolute accuracy improvements on audio-visual QA tasks with human speech. In addition, video-SALMONN demonstrates remarkable video comprehension and reasoning abilities on tasks that are unprecedented by other av-LLMs. Our training code and model checkpoints are available at https://github.com/bytedance/SALMONN/} }
Endnote
%0 Conference Paper %T video-SALMONN: Speech-Enhanced Audio-Visual Large Language Models %A Guangzhi Sun %A Wenyi Yu %A Changli Tang %A Xianzhao Chen %A Tian Tan %A Wei Li %A Lu Lu %A Zejun Ma %A Yuxuan Wang %A Chao Zhang %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-sun24l %I PMLR %P 47198--47217 %U https://proceedings.mlr.press/v235/sun24l.html %V 235 %X Speech understanding as an element of the more generic video understanding using audio-visual large language models (av-LLMs) is a crucial yet understudied aspect. This paper proposes video-SALMONN, a single end-to-end av-LLM for video processing, which can understand not only visual frame sequences, audio events and music, but speech as well. To obtain fine-grained temporal information required by speech understanding, while keeping efficient for other video elements, this paper proposes a novel multi-resolution causal Q-Former (MRC Q-Former) structure to connect pre-trained audio-visual encoders and the backbone large language model. Moreover, dedicated training approaches including the diversity loss and the unpaired audio-visual mixed training scheme are proposed to avoid frames or modality dominance. On the introduced audio-visual evaluation benchmark, video-SALMONN achieves more than 25% absolute accuracy improvements on the video-QA task and over 30% absolute accuracy improvements on audio-visual QA tasks with human speech. In addition, video-SALMONN demonstrates remarkable video comprehension and reasoning abilities on tasks that are unprecedented by other av-LLMs. Our training code and model checkpoints are available at https://github.com/bytedance/SALMONN/
APA
Sun, G., Yu, W., Tang, C., Chen, X., Tan, T., Li, W., Lu, L., Ma, Z., Wang, Y. & Zhang, C.. (2024). video-SALMONN: Speech-Enhanced Audio-Visual Large Language Models. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:47198-47217 Available from https://proceedings.mlr.press/v235/sun24l.html.

Related Material