Multistage linguistic conditioning of convolutional layers for speech emotion recognition
arXiv:2110.06650 · doi:10.3389/fcomp.2023.1072479
Abstract
In this contribution, we investigate the effectiveness of deep fusion of text and audio features for categorical and dimensional speech emotion recognition (SER). We propose a novel, multistage fusion method where the two information streams are integrated in several layers of a deep neural network (DNN), and contrast it with a single-stage one where the streams are merged in a single point. Both methods depend on extracting summary linguistic embeddings from a pre-trained BERT model, and conditioning one or more intermediate representations of a convolutional model operating on log-Mel spectrograms. Experiments on the MSP-Podcast and IEMOCAP datasets demonstrate that the two fusion methods clearly outperform a shallow (late) fusion baseline and their unimodal constituents, both in terms of quantitative performance and qualitative behaviour. Overall, our multistage fusion shows better quantitative performance, surpassing alternatives on most of our evaluations. This illustrates the potential of multistage fusion in better assimilating text and audio information.
References in corpus (5)
- WaveNet: A Generative Model for Raw Audio
- vq-wav2vec: Self-Supervised Learning of Discrete Speech Representations
- Fusion approaches for emotion recognition from speech using acoustic and text-based features
- On the Impact of Word Error Rate on Acoustic-Linguistic Speech Emotion Recognition: An Update for the Deep Learning Era
- The MuSe 2021 Multimodal Sentiment Analysis Challenge: Sentiment, Emotion, Physiological-Emotion, and Stress