Start a Transcription and Captioning Service

A transcription and captioning service converts audio and video into accurate text for medical, legal, and media clients, charging per audio minute or per project. People assume AI killed it, but legal, medical, and accessibility work still requires accurate human review and pays for it. You pick one paying lane (legal, medical, or ADA captioning), pass a platform test or take direct clients, and set per-minute rates ($1 to $3 for direct clients, more for verbatim or rush). AI now does the first pass, so you edit for accuracy and roughly double your output while keeping the human-quality guarantee. Startup runs under $500, margins run 60 to 75 percent, and it is a flexible, home-based, beginner-friendly business.

Accuracy still matters where the stakes are high: a legal transcript, a medical record, or ADA-compliant captions cannot be wrong, so clients pay humans to guarantee quality that raw AI cannot. The reframe most people miss: AI is your tool, not your competitor, because running audio through speech-to-text and then editing lets you keep the accuracy guarantee while doubling throughput. Podcasters, law firms, universities, and video creators need this every week, so one retainer client beats fifty marketplace gigs. Low startup cost and remote work make it accessible, and specializing in a paying lane keeps you out of the race-to-the-bottom general-transcription market.

Because AI transcription is everywhere, people assume the work is gone, and general transcription has indeed become a low-paid commodity. The overlooked reality is that legal, medical, and accessibility captioning still demand human-verified accuracy and pay professional rates. Most people never specialize into those lanes or build direct client relationships, so the transcriptionist who does, and who uses AI as a first pass rather than fearing it, finds steady, better-paid work.