This is the respository of baseline systems for Multimodal Information Based Speech Processing (MISP) 2022 Challenge. In this repository, we provide a baseline system for Track 1 - Audio-Visual Speaker Diarization (AVSD), please refer to the track1_AVSD folder for details and a baseline system for Track 2 - Audio-Visual Diarization and Recognition (AVDR), please refer to the track2_AVDR folder for details.