AMI Meeting Corpus - Dialogue and Multi-Speaker Conversations

Multi-speaker meeting recordings, useful for conversation modeling.

About Dataset

The AMI Meeting Corpus is a multi-modal dataset capturing real and simulated meetings, including audio recordings, transcriptions, video, and annotations. Developed by the AMI Consortium, it reflects natural group interactions, discussions, and collaborative decision-making scenarios.

Purpose of Dataset

The AMI Corpus is used for research in speech recognition, speaker diarization, meeting summarization, and multimodal interaction analysis. It supports development of systems that understand conversational dynamics, turn-taking, and collaborative speech in real-world meeting environments.