SoniaSolutions/AMI-universal

AMI通用数据集是一个用于评估自动语音识别（ASR）、语音活动检测（VAD）和说话者标注的三合一数据集。它包含了音频文件、时间戳信息、说话者信息和转录文本。数据集由两个其他数据集组合而成，以确保涵盖所有三个任务的必要信息。数据集分为验证集和测试集。

The AMI Universal dataset is a combination for evaluating Automatic Speech Recognition (ASR), Voice Activity Detection (VAD), and Speaker Diarization. It includes audio files, timestamp information, speaker information, and transcripts. The dataset is compiled from two other datasets to ensure it covers all necessary information for the three tasks. The dataset is split into validation and test sets.