XTREME-S project

该数据集包含102种语言的语音数据，涵盖了索马里语、斯瓦希里语、卢奥语和肯尼亚的坎巴语等。其规模宏大，不仅包含了多种语言，还特别关注了肯尼亚的语言。该数据集的任务是面向多语言语音处理。This dataset contains speech data across 102 languages, including Somali, Swahili, Luo, and Kamba of Kenya, among others. As a large-scale resource, it not only covers a broad spectrum of languages but also places special focus on languages native to Kenya. This dataset is tailored for multilingual speech processing tasks.