The AssemblyAI MCP server is an implementation of the Model Context Protocol that connects your AI agent and assistants like Claude, Cursor, etc directly to your AssemblyAI account. It provides structured and secure access to your media and transcription workflows, so your agent can upload recordings, submit transcriptions, retrieve results, search transcripts, and export subtitles on your behalf.
- Media upload and transcription: Have your agent upload an audio or video file, or submit remotely hosted media for transcription.
- Transcript retrieval and review: Let the agent check transcription progress, retrieve completed text, and list recent transcripts from your account.
- Detailed transcript segments: Direct your agent to fetch sentence or paragraph segments with timestamps, confidence scores, and individual words.
- Transcript search and subtitle export: Instruct your agent to find words or phrases with occurrence timestamps, or export completed transcripts as SRT or WebVTT subtitles.
- Redacted audio and transcript cleanup: Have your agent retrieve available PII-redacted audio or permanently delete stored transcript content you no longer need.