JSSS

JSSS数据集是由东京大学信息科学与技术研究生院和东京都市大学系统设计研究生院联合开发的日语语音数据集，旨在支持语音摘要和简化任务。该数据集包含约8小时的语音数据，涵盖了时长约束的文本到语音摘要和语音风格简化等多种任务。数据集的创建过程包括高质量的录音设置和专业的语音录制，确保了数据的准确性和可用性。JSSS数据集的应用领域广泛，主要用于提升非母语者的语音理解能力和在时间限制下的信息传递效率。

The JSSS dataset is a Japanese speech corpus jointly developed by the Graduate School of Information Science and Technology at The University of Tokyo and the Graduate School of System Design at Tokyo City University, with the core objective of supporting speech summarization and simplification tasks. This dataset contains approximately 8 hours of speech data, encompassing multiple tasks including duration-constrained text-to-speech summarization and speech style simplification. The construction process of the dataset features high-quality recording setup and professional speech recording, ensuring the accuracy and usability of the data. The JSSS dataset has broad application scenarios, which are mainly utilized to enhance the speech comprehension ability of non-native speakers and improve the efficiency of information transmission under time constraints.