Media, methods, and systems are disclosed for documenting synchronous multimedia collaboration sessions in a group-based communication system. A live synchronous multimedia collaboration session is detected. In response to the detecting, a first audio segment is received and transcribed into a first text segment. A collaboration thread associated with the live synchronous multimedia collaboration session is generated. In real time, a first message is posted based on the first transcribed text segment to the collaboration thread associated with the live synchronous multimedia collaboration session. Next, a second audio segment is received and transcribed into a second text segment. Finally, in real time and during the live synchronous multimedia collaboration session, a second message is posted to the collaboration thread based on the second transcribed text segment.