Interview notes & audio transcription integration
Prompt detail
Generate a user interview record based on the original audio transcription of the user interview and the draft interview record. The original audio transcription can be found in the document {{@Document Name}}, and the draft interview record is in the document {{@Document Name}}. Prioritize the content of the draft interview record, and revise it using the information from the original audio transcription to make the content more accurate and comprehensive. Keep the outline structure and questions in the draft interview record unchanged — the outline and questions themselves must remain as they are.
What the prompt does
Reconciling raw audio transcriptions with handwritten interview notes is tedious and error-prone. Key quotes get lost, paraphrased answers drift from what users actually said, and important nuances slip through the cracks.
With remio, you point to both your audio transcription document and your draft interview record, and it intelligently merges them — preserving your original outline and question structure while filling in gaps, correcting inaccuracies, and enriching answers with verbatim details from the transcription. The output is a comprehensive, polished interview record ready for stakeholder review.
When to use
• After conducting a customer discovery interview where you took quick notes but need a complete, accurate record for the product team.
• When your notetaker's draft misses critical user quotes or sentiment that the audio transcription captured.
• Before synthesizing findings across multiple interviews — ensuring each individual record is accurate first.
• When preparing interview highlights for a research repository and you need verbatim accuracy for compliance or credibility.
• After a usability testing session where the moderator's notes need reconciliation with what participants actually said on recording.
Prompt examples
Input: Generate a user interview record based on the original audio transcription of the user interview and the draft interview record. The original audio transcription can be found in the document {{@Q3 Onboarding Research - Sarah Chen Transcript}}, and the draft interview record is in the document {{@Q3 Onboarding Research - Sarah Chen Draft Notes}}. Prioritize the content of the draft interview record, and revise it using the information from the original audio transcription to make the content more accurate and comprehensive. Keep the outline structure and questions in the draft interview record unchanged — the outline and questions themselves must remain as they are.
remio responds:
Here is your revised interview record for Sarah Chen, following the exact outline from your draft notes. Under "Q2: What was your first impression of the onboarding flow?" I've replaced your paraphrase with her actual response: "Honestly, I felt overwhelmed — there were too many steps before I could even see the dashboard." I also added a detail your draft missed under Q5 — Sarah mentioned she almost churned during week two because she couldn't find the integrations page. The sentiment tags you included in the draft are preserved, and I've flagged three spots where the transcription contradicts or significantly expands on your original notes.
Tip 1: Ensure your draft interview record has clearly labeled questions and sections — remio uses that structure as the skeleton and will not modify it, so well-organized drafts yield cleaner results.
Tip 2: If your audio transcription includes multiple speakers, label them in the document beforehand so remio can accurately attribute quotes to the interviewee versus the interviewer.
Tip 3: After generating the merged record, ask remio to highlight direct quotes versus paraphrased content for easier downstream analysis.
More tips
remio meets all your needs about knowledge
Learn more features
Capture resources from websites, local folders & files.
Get instant, reliable answers from your entire knowledge base.
Your working assistant, on desktop, browser, and mobile.
Unlimited free recording and transcription, no-bot.