Itbanque/ScreenTalk_JA2ZH-XS

ScreenTalk_JA2ZH-XS是一个包含日语语音和简体中文翻译文本对的数据集，适用于训练和评估语音翻译和多语种语音理解模型。该数据集由真实世界的日本电影和电视剧中的对话组成，包含大约10,000个样本，总时长约30小时。ScreenTalk_JA2ZH-XS is a dataset consisting of paired Japanese speech and Simplified Chinese translated text, designed for training and evaluating speech translation and multilingual speech understanding models. The dataset comprises dialogues extracted from real-world Japanese movies and TV shows, including approximately 10,000 samples with a total duration of about 30 hours.