Every order goes through the same four steps. Software does the fast part and a person does the careful part.
1. You upload your file
Choose an audio or video file and a turnaround. Your browser reads the file’s length and shows the exact price before you pay. The file uploads straight to private cloud storage, so large recordings are fine.
You can list names, company terms, acronyms or jargon that appear in the recording. We use the list twice: to guide the speech recognition, and as a checklist for the reviewer.
2. Speech recognition makes a draft
Once your payment clears, the recording goes to our speech recognition provider. It returns a draft with timestamps, a label for each speaker, and a confidence score for every word.
3. A person reviews it
A reviewer opens the draft beside the audio. Words the software was unsure of are highlighted, so the reviewer listens closely there. They:
- correct misheard words, punctuation and line breaks
- replace “Speaker A” and “Speaker B” with real names or roles
- tick off every term on your list against the audio
- fix where one speaker’s turn ends and the next begins
A second check then approves the transcript before it is released.
4. You download it
We email you as soon as it is ready. From your orders page you can download:
- Word and PDF, with speaker names and timestamps
- Plain text
- SRT and VTT caption files for video players, YouTube and editing software
If anything needs fixing, reply to the delivery email within 7 days and we will correct it at no charge.