prosodyntax/vxp-perspeak-withprom-filtered-fixed

这是一个包含多种语言的数据集，每种语言都有单元、单元时长、电话号码、电话号码时长、多标记词、联合发音、词性、头词、依赖关系、UD ID、句子 ID、说话者、突出度强度、边界强度和块标签等特征。数据集为训练目的划分了训练集，并提供了每种语言训练集的大小和示例数量。

This is a dataset containing multiple languages, each with features such as unit, unit duration, phone numbers, phone number duration, multitoken words, joint pronunciation, parts of speech, head words, dependency relations, UD ID, sentence ID, speaker, prominence strength, boundary strength, and chunk labels. The dataset is divided into training sets for each language, and the size and number of examples of each training set are provided.