Parameter Efficient Finetuning for Speech Emotion Recognition and Domain Adaptation
This work addresses the challenge of efficient and effective model adaptation for speech emotion recognition, particularly in resource-limited settings, though it is incremental as it builds on existing PEFT techniques.
The paper tackles the problem of resource-intensive and overfitting-prone full finetuning of large pre-trained models for speech emotion recognition by investigating parameter-efficient finetuning methods, showing that combining these methods surpasses full finetuning with significantly fewer trainable parameters and proposing a two-stage adaptation strategy to improve performance on natural emotional expressions.
Foundation models have shown superior performance for speech emotion recognition (SER). However, given the limited data in emotion corpora, finetuning all parameters of large pre-trained models for SER can be both resource-intensive and susceptible to overfitting. This paper investigates parameter-efficient finetuning (PEFT) for SER. Various PEFT adaptors are systematically studied for both classification of discrete emotion categories and prediction of dimensional emotional attributes. The results demonstrate that the combination of PEFT methods surpasses full finetuning with a significant reduction in the number of trainable parameters. Furthermore, a two-stage adaptation strategy is proposed to adapt models trained on acted emotion data, which is more readily available, to make the model more adept at capturing natural emotional expressions. Both intra- and cross-corpus experiments validate the efficacy of the proposed approach in enhancing the performance on both the source and target domains.