Codyfederer/tr-combined

Tr_combined是一个包含894个原始数据集合并而成的语音数据集，共有221531个语音段。该数据集支持土耳其语，并标注了说话者的情感，包括快乐、愤怒、中性和悲伤。每个样本包括音频文件、文本转录、说话者ID、语言、情感、原始数据集名称、原始文件名以及音频片段的开始时间和持续时间。

Tr_combined is a speech dataset merged from 894 original datasets, containing a total of 221,531 audio segments. The dataset supports Turkish language and is annotated with speakers emotions, including happy, angry, neutral, and sad. Each sample includes an audio file, text transcription, speaker ID, language, emotion, name of the original dataset, original filename, as well as the start time and duration of the audio segment.