TAVT: Towards Transferable Audio-Visual Text Generation

Wang Lin, Tao Jin, Wenwen Pan, Linjun Li, Xize Cheng, Ye Wang, Zhou Zhao


Abstract
Audio-visual text generation aims to understand multi-modality contents and translate them into texts. Although various transfer learning techniques of text generation have been proposed, they focused on uni-modal analysis (e.g. text-to-text, visual-to-text) and lack consideration of multi-modal content and cross-modal relation. Motivated by the fact that humans can recognize the timbre of the same low-level concepts (e.g., footstep, rainfall, and laughing), even in different visual conditions, we aim to mitigate the domain discrepancies by audio-visual correlation. In this paper, we propose a novel Transferable Audio-Visual Text Generation framework, named TAVT, which consists of two key components: Audio-Visual Meta-Mapper (AVMM) and Dual Counterfactual Contrastive Learning (DCCL). (1) AVMM first introduces a universal auditory semantic space and drifts the domain-invariant low-level concepts into visual prefixes. Then the reconstruct-based learning encourages the AVMM to learn “which pixels belong to the same sound” and achieve audio-enhanced visual prefix. The well-trained AVMM can be further applied to uni-modal setting. (2) Furthermore, DCCL leverages the destructive counterfactual transformations to provide cross-modal constraints for AVMM from the perspective of feature distribution and text generation. (3) The experimental results show that TAVT outperforms the state-of-the-art methods across multiple domains (cross-datasets, cross-categories) and various modal settings (uni-modal, multi-modal).
Anthology ID:
2023.acl-long.836
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
14983–14999
Language:
URL:
https://aclanthology.org/2023.acl-long.836
DOI:
10.18653/v1/2023.acl-long.836
Bibkey:
Cite (ACL):
Wang Lin, Tao Jin, Wenwen Pan, Linjun Li, Xize Cheng, Ye Wang, and Zhou Zhao. 2023. TAVT: Towards Transferable Audio-Visual Text Generation. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 14983–14999, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
TAVT: Towards Transferable Audio-Visual Text Generation (Lin et al., ACL 2023)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingest-2024-clasp/2023.acl-long.836.pdf
Video:
 https://preview.aclanthology.org/ingest-2024-clasp/2023.acl-long.836.mp4