用压缩版模型实现低功耗手机情绪识别，跨语料训练提升稳定性。
Distilled HuBERT for Mobile Speech Emotion Recognition: A Cross-Corpus Validation Study
- 将HuBERT模型蒸馏量化至8比特，参数减少92%，仅23MB大小。
- 跨语料训练使准确率提升1.2%，方差降低32%，中性类识别改善5.4%。
- 适合移动端部署，尤其在资源受限场景下保持情绪判别鲁棒性。
语音情绪识别（SER）在移动端有广泛应用潜力，但主流Transformer模型计算开销大。本文提出基于DistilHuBERT的轻量级SER系统，该模型经蒸馏与8比特量化，相比全尺寸Wav2Vec 2.0减少约92%参数，仍保持良好性能。在IEMOCAP数据集上采用5折留一会话交叉验证以确保说话人无关性，并引入CREMA-D数据进行跨语料训练，使加权准确率提升1.2%，宏平均F1提升1.4%，跨折叠方差降低32%，其中中性类F1提升5.4%。模型最终达到61.4%无加权准确率，仅23MB大小，约为全尺寸基线的91%。在RAVDESS上的跨语料评估显示，戏剧化表演导致情绪预测按唤醒度聚集：快乐常误判为愤怒，悲伤常误判为中性，因演员追求清晰度而非细微差异。尽管整体准确率降至46.64%，模型对愤怒唤醒度的召回率达99%，中性为55%，悲伤为27%。结果表明该方案在模型大小与精度间达成帕累托最优，适用于资源受限的移动设备。
原文摘要 · Abstract (English)
Speech Emotion Recognition (SER) has significant potential for mobile applications, yet deployment remains constrained by the computational demands of state-of-the-art transformer architectures. This paper presents a mobile-efficient SER system based on DistilHuBERT, a distilled and 8-bit quantized transformer that achieves approximately 92% parameter reduction compared to full-scale Wav2Vec 2.0 models while maintaining competitive accuracy. We conduct a rigorous 5-fold Leave-One-Session-Out (LOSO) cross-validation on the IEMOCAP dataset to ensure speaker independence, augmented with cross-corpus training on CREMA-D to enhance generalization. Cross-corpus training with CREMA-D yields a 1.2% improvement in Weighted Accuracy, a 1.4% gain in Macro F1-score, and a 32% reduction in cross-fold variance, with the Neutral class showing the most substantial benefit at 5.4% F1-score improvement. Our approach achieves an Unweighted Accuracy of 61.4% with a quantized model footprint of only 23 MB, representing approximately 91% of the Unweighted Accuracy of a full-scale baseline. Cross-corpus evaluation on RAVDESS reveals that the theatrical nature of acted emotions causes predictions to cluster by arousal level rather than by specific emotion categories - happiness predictions systematically bleed into anger predictions, and sadness predictions bleed into neutral predictions, due to acoustic saturation when actors prioritize clarity over subtlety. Despite this theatricality effect reducing overall RAVDESS accuracy to 46.64%, the model maintains robust arousal detection with 99% recall for anger, 55% recall for neutral, and 27% recall for sadness. These findings demonstrate a Pareto-optimal tradeoff between model size and accuracy, enabling practical affect recognition on resource-constrained mobile devices.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。