Active Learning Based Fine-Tuning Framework for Speech Emotion Recognition
Not on the map. Why?
Abstract (arXiv)
Speech emotion recognition (SER) has drawn increasing attention for its applications in human-machine interaction. However, existing SER methods ignore the information gap between the pre-training speech recognition task and the downstream SER task, leading to sub-optimal performance. Moreover, they require much time to fine-tune on each specific speech dataset, restricting their effectiveness in real-world scenes with large-scale noisy data. To address these issues, we propose an active learning (AL) based Fine-Tuning framework for SER that leverages task adaptation pre-training (TAPT) and AL methods to enhance performance and efficiency. Specifically, we first use TAPT to minimize the information gap between the pre-training and the downstream task. Then, AL methods are used to iteratively select a subset of the most informative and diverse samples for fine-tuning, reducing time consumption. Experiments demonstrate that using only 20%pt. samples improves 8.45%pt. accuracy and reduces 79%pt. time consumption.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 25 of its 50 references that are in the map. First posted on arXiv 30 Sep 2023.
Builds on
25 of 50 references.
References · 25
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
- 2020wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representationscited by 3,853
- 2020Don't Stop Pretraining: Adapt Language Models to Domains and Taskscited by 1,548
- 2020Graph Contrastive Learning with Augmentationscited by 1,203
- 2022data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Languagecited by 718
- 2021XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scalecited by 589
- 2019BatchBALD: Efficient and Diverse Batch Acquisition for Deep Bayesian Active Learningcited by 428
- 2022Dawn of the transformer era in speech emotion recognition: closing the valence gapcited by 155
- 2021Active Learning by Acquiring Contrastive Examplescited by 154
- 2020Cold-start Active Learning through Self-supervised Language Modelingcited by 128
- 2018The Emotional Voices Database: Towards Controlling the Emotion Dimension in Voice Generation Systemscited by 61
13 more
- 2022Speech Emotion Recognition using Self-Supervised Featurescited by 52
- 2021Exploring Wav2vec 2.0 fine-tuning for improved speech emotion recognitioncited by 47
- 2019ShEMO -- A Large-Scale Validated Database for Persian Speech Emotion Detectioncited by 21
- 2021LSSED: a large-scale dataset and benchmark for speech emotion recognitioncited by 17
- 2023Vesper: A Compact and Effective Pretrained Model for Speech Emotion Recognitioncited by 16
- 2022Speech Emotion Recognition using Supervised Deep Recurrent System for Mental Health Monitoringcited by 15
- 2021Improved Speech Emotion Recognition using Transfer Learning and Spectrogram Augmentationcited by 12
- 2022Sentiment-Aware Automatic Speech Recognition pre-training for enhanced Speech Emotion Recognitioncited by 8
- 2022Learning Speech Emotion Representations in the Quaternion Domaincited by 7
- 2022Fast Yet Effective Speech Emotion Recognition with Self-distillationcited by 7
- 2022Speech Emotion Recognition with Global-Aware Fusion on Multi-scale Feature Representationcited by 4
- 2021Unsupervised low-rank representations for speech emotion recognitioncited by 3
- 2022Domain Specific Wav2vec 2.0 Fine-tuning For The SE&R 2022 Challengecited by 3
Built on by
Citing papers per year: 2024: 2.
2 papers.