CTAL: Pre-training Cross-modal Transformer for Audio-and-Language Representations

Hang Li, Wenbiao Ding, Yu Kang, Tianqiao Liu, Zhongqin Wu, Zitao Liu


Abstract
Existing audio-language task-specific predictive approaches focus on building complicated late-fusion mechanisms. However, these models are facing challenges of overfitting with limited labels and low model generalization abilities. In this paper, we present a Cross-modal Transformer for Audio-and-Language, i.e., CTAL, which aims to learn the intra-modality and inter-modality connections between audio and language through two proxy tasks on a large amount of audio-and-language pairs: masked language modeling and masked cross-modal acoustic modeling. After fine-tuning our pre-trained model on multiple downstream audio-and-language tasks, we observe significant improvements across various tasks, such as, emotion classification, sentiment analysis, and speaker verification. On this basis, we further propose a specially-designed fusion mechanism that can be used in fine-tuning phase, which allows our pre-trained model to achieve better performance. Lastly, we demonstrate detailed ablation studies to prove that both our novel cross-modality fusion component and audio-language pre-training methods significantly contribute to the promising results. The code and pre-trained models are available at https://github.com/tal-ai/CTAL_EMNLP2021.
Anthology ID:
2021.emnlp-main.323
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Editors:
Marie-Francine Moens, Xuanjing Huang, Lucia Specia, Scott Wen-tau Yih
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
3966–3977
Language:
URL:
https://aclanthology.org/2021.emnlp-main.323
DOI:
10.18653/v1/2021.emnlp-main.323
Bibkey:
Cite (ACL):
Hang Li, Wenbiao Ding, Yu Kang, Tianqiao Liu, Zhongqin Wu, and Zitao Liu. 2021. CTAL: Pre-training Cross-modal Transformer for Audio-and-Language Representations. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 3966–3977, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
CTAL: Pre-training Cross-modal Transformer for Audio-and-Language Representations (Li et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.323.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.323.mp4
Code
 ydkwim/ctal
Data
IEMOCAPLibriSpeech