Wav2small: Distilling Wav2vec2 To 72K Parameters For Low-resource Speech Emotion Recognition

Abstract

Speech Emotion Recognition (SER) needs high computational resources to overcome the challenge of substantial annotator disagreement. Today SER is shifting towards dimensional annotations of arousal, dominance, and valence (A/D/V). Universal metrics as the L2 distance prove unsuitable for evaluating A/D/V accuracy due to non converging consensus of annotator opinions. However, Concordance Correlation Coefficient (CCC) arose as an alternative metric for A/D/V where a model's output is evaluated to match a whole dataset's CCC rather than L2 distances of individual audios. Recent studies have shown that wav2vec2 / wavLM architectures outputing a float value for each A/D/V dimension achieve today's State-of-the-art (Sota) CCC on A/D/V. The Wav2Vec2.0 / WavLM family has a high computational footprint, but training small models using human annotations has been unsuccessful. In this paper we use a large Transformer Sota A/D/V model as Teacher/Annotator to train 5 student models: 4 MobileNets a

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keykounadisbastian2024wav2small

Related papers

• Dawn Of The Transformer Era In Speech Emotion Recognition: Closing The Valence Gap (2022)18.59
• Distilled Hubert For Mobile Speech Emotion Recognition: A Cross-corpus Validation Study (2025)0.00
• Exploring Wav2vec 2.0 Fine-tuning For Improved Speech Emotion Recognition (2021)15.67
• Sigwavnet: Learning Multiresolution Signal Wavelet Network For Speech Emotion Recognition (2025)8.48
• Speaker Emotion Recognition: Leveraging Self-supervised Models For Feature Extraction Using Wav2vec2 And Hubert (2024)0.00
• On The Use Of Self-supervised Pre-trained Acoustic And Linguistic Features For Continuous Speech Emotion Recognition (2020)11.85
• Fine-tuning Wav2vec2 For Speaker Recognition (2021)18.88
• Continuous Metric Learning For Transferable Speech Emotion Recognition And Embedding Across Low-resource Languages (2022)0.00