Journal
IEEE MULTIMEDIA
Volume 29, Issue 2, Pages 94-103Publisher
IEEE COMPUTER SOC
DOI: 10.1109/MMUL.2022.3161411
Keywords
Acoustics; Feature extraction; Transformers; Emotion recognition; Linguistics; Data mining; Speech recognition
Categories
Funding
- National Key R&D Program of China [2018YFB1305200]
- National Natural Science Foundation of China [62176182, 61976216]
- Tianjin Municipal Science and Technology Project [19ZXZNGX00030]
Ask authors/readers for more resources
This article proposes an implicitly aligned multimodal transformer fusion framework based on acoustic features and text information for emotion recognition. The model allows two modalities to guide and complement each other, and uses weighted fusion to control the contributions of different modalities, thereby obtaining more complementary emotional representations. Experiments have shown that this method outperforms baseline methods.
People usually express emotions through paralinguistic and linguistic information in speech. How to effectively integrate linguistic and paralinguistic information for emotion recognition is a challenge. Previous studies have adopted the bidirectional long short-term memory (BLSTM) network to extract acoustic and lexical representations followed by a concatenate layer, and this has become a common method. However, the interaction and influence between different modalities are difficult to promote using simple feature fusion for each sentence. In this article, we propose an implicitly aligned multimodal transformer fusion (IA-MMTF) framework based on acoustic features and text information. This model enables the two modalities to guide and complement each other when learning emotional representations. Thereafter, the weighed fusion is used to control the contributions of different modalities. Thus, we can obtain more complementary emotional representations. Experiments on the interactive emotional dyadic motion capture (IEMOCAP) database and multimodal emotionlines dataset (MELD) show that the proposed method outperforms the baseline BLSTM-based method.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available