Modulated Fusion using Transformer for Linguistic-Acoustic Emotion Recognition
arXiv:2010.02057
Abstract
This paper aims to bring a new lightweight yet powerful solution for the task of Emotion Recognition and Sentiment Analysis. Our motivation is to propose two architectures based on Transformers and modulation that combine the linguistic and acoustic inputs from a wide range of datasets to challenge, and sometimes surpass, the state-of-the-art in the field. To demonstrate the efficiency of our models, we carefully evaluate their performances on the IEMOCAP, MOSI, MOSEI and MELD dataset. The experiments can be directly replicated and the code is fully open for future researches.
arXiv commentEMNLP 2020 workshop: NLP Beyond Text (NLPBT)
References in corpus (6)
- Adam: A Method for Stochastic Optimization
- MOSI: Multimodal Corpus of Sentiment Intensity and Subjectivity Analysis in Online Opinion Videos
- Multilogue-Net: A Context Aware RNN for Multi-modal Emotion Detection and Sentiment Analysis in Conversation
- Deep Modular Co-Attention Networks for Visual Question Answering
- Multimodal Speech Emotion Recognition and Ambiguity Resolution
- Seq2Seq2Sentiment: Multimodal Sequence to Sequence Models for Sentiment Analysis