MediaSpeech

该数据集包含了法语、土耳其语和西班牙语的媒体语音，最初是为了测试自动语音识别系统的性能而构建的。它包含了三种设定语言中的媒体语音数据，每种语言都有10小时的语音资料。此外，该数据集的任务是进行语音语言识别。

This dataset encompasses media speech in French, Turkish and Spanish, and was initially constructed to test the performance of automatic speech recognition (ASR) systems. It contains media speech data across the three target languages, with 10 hours of speech material for each language. Additionally, the task associated with this dataset is spoken language identification (SLI).