SCRIBE

该数据集名为SCRIBE，是一个包含英式英语朗读和自发语音的小型语料库。它由音素丰富且简洁的句子组成，非常适合用于评估音素识别系统。该数据集包括200个音素丰富的句子和460个音素紧凑的句子，且音频文件已针对音素标注进行了处理，任务旨在进行音素识别和对齐。

Named SCRIBE, this dataset is a small corpus of British English read and spontaneous speech. It consists of sentences that are phonemically rich and concise, making it ideal for evaluating phoneme recognition systems. The dataset includes 200 such phonemically rich and concise sentences, plus 460 phonemically compact sentences. All audio files have been processed to support phoneme annotation, and the core task of this dataset is phoneme recognition and alignment.