We propose a novel multi-task pre-training method for Speech Emotion
Recognition (SER). We pre-train SER model simultaneously on Automatic Speech
Recognition (ASR) and sentiment classification tasks to make the acoustic ASR
model more ``emotion aware''. We generate targets for the sentiment
classification using text-to-sentiment model trained on publicly available
data. Finally, we fine-tune the acoustic ASR on emotion annotated speech data.
We evaluated the proposed approach on the MSP-Podcast dataset, where we
achieved the best reported concordance correlation coefficient (CCC) of 0.41
for valence prediction.

提出了一种新颖的多任务预训练方法，用于语音情感识别，在同时进行自动语音识别和情感分类任务的情况下对 SER 模型进行预训练，生成目标情感分类的方法，通过公开数据训练的文本情感模型，将声学 ASR 模型更加 “情感感知”，最终在情感注释语音数据上进行了微调，达到了 MSP-Podcast 数据集中价值预测最佳的一致性相关系数（CCC）0.41。