Summary
ACTA is an internal meeting platform built at OSIX Tech.
It handles transcription, speaker diarization, summary generation, and automatic task assignment per team member. The goal is simple: meetings should produce useful operational output without relying on someone to manually clean up notes afterward.
Problem
Meetings often contain decisions and tasks, but the follow-through gets lost because notes are inconsistent, ownership is unclear, or nobody wants to do the cleanup work after the call.
Approach
ACTA uses voice AI to transcribe meetings and identify speakers, then uses LLMs to structure the result into summaries and task assignments.
Outcome
ACTA turns meetings into a cleaner operational record: what happened, who said what, and what needs to happen next.