EmoFilm

该数据集包含了用于训练和测试语音情感识别（SER）模型的情感语音剪辑片段。在遵循五折交叉验证方法的前提下，该数据集中的85%被用于训练模型。该任务的目的是实现语音情感识别。

This dataset contains emotional speech clips for training and testing speech emotion recognition (SER) models. Following the 5-fold cross-validation methodology, 85% of the dataset is allocated for model training. The goal of this task is speech emotion recognition.