@inproceedings{lin-etal-2023-tavt,
title = "{TAVT}: Towards Transferable Audio-Visual Text Generation",
author = "Lin, Wang and
Jin, Tao and
Pan, Wenwen and
Li, Linjun and
Cheng, Xize and
Wang, Ye and
Zhao, Zhou",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.836",
doi = "10.18653/v1/2023.acl-long.836",
pages = "14983--14999",
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).",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lin-etal-2023-tavt">
<titleInfo>
<title>TAVT: Towards Transferable Audio-Visual Text Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wang</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tao</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenwen</namePart>
<namePart type="family">Pan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Linjun</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xize</namePart>
<namePart type="family">Cheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ye</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhou</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<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).</abstract>
<identifier type="citekey">lin-etal-2023-tavt</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.836</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.836</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>14983</start>
<end>14999</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TAVT: Towards Transferable Audio-Visual Text Generation
%A Lin, Wang
%A Jin, Tao
%A Pan, Wenwen
%A Li, Linjun
%A Cheng, Xize
%A Wang, Ye
%A Zhao, Zhou
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F lin-etal-2023-tavt
%X 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).
%R 10.18653/v1/2023.acl-long.836
%U https://aclanthology.org/2023.acl-long.836
%U https://doi.org/10.18653/v1/2023.acl-long.836
%P 14983-14999
Markdown (Informal)
[TAVT: Towards Transferable Audio-Visual Text Generation](https://aclanthology.org/2023.acl-long.836) (Lin et al., ACL 2023)
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.