phonemetransformers/IPA-CHILDES

IPA-CHILDES数据集包含了从儿童语言数据库CHILDES下载并转换为音素表示形式的语句。这个数据集经过预处理，保留下许多原始CHILDES数据库中的列，并添加了几个关键列，以方便进行语言模型训练。数据集覆盖了多种语言，每种语言都有详细的统计数据，如说话者数量、语句数量等。

The IPA-CHILDES dataset contains utterances downloaded from CHILDES, which have been pre-processed and converted into a phonemic representation. This dataset includes multiple language versions, each phonemized with a specific language code. It also includes key columns such as pre-processed orthographic transcription, phonemic transcription, character-split utterance, and an indicator of whether the utterance was spoken by a child. The dataset is sorted by child age and can be used to limit the training data. Each language has a detailed description including data sources, number of speakers, number of utterances, number of words, number of phonemes, and the percentage of child utterances.