Interviews
Interview transcription
Audicite transcribes interviews with the interviewer and each guest labelled and every word timed, so a quote can be played back before it is used. Record live in person or on a call, or upload the recording. Quotes it suggests are the exact words, checked against the transcript.
Updated by Audicite
Who it is for
- Journalists and writers
- Qualitative researchers and user researchers
- Recruiters and HR teams documenting interviews
What Audicite does
- Labels speakers, even in quick back-and-forth
- Suggests verbatim quotes and lets you mark your own, each playable
- Hides names, emails and numbers in exports (redaction)
- Lets you verify the transcript paragraph by paragraph against the audio
Key facts
| Speaker labels | Yes, automatic; rename a speaker once and it changes everywhere |
|---|---|
| Timestamps | Every word is timed; click a word to play from it |
| Quotes | Verbatim, checked word for word against the transcript |
| Privacy | Redaction in exports, retention controls, delete any time |
| Export formats | Word (DOCX, with sources), PDF (with sources), Plain text, SRT captions, WebVTT captions, EBU STL subtitles, EDL (rough cut), CSV, HTML, XML, JSON |
| Price | Free trial (30 minutes, no card), then $14.99 per person a month |
How it works
- 01Record or upload. Live on your phone or laptop, or upload the recording.
- 02Name the speakers. Rename “Speaker 2” once and it changes everywhere.
- 03Pull quotes. Use the suggested quotes or mark your own; play each before using it.
- 04Export. Word or PDF with timestamps, or plain text.
Examples
A feature interview
Two hours of conversation, searchable, with the strongest quotes ready.
User research
Twelve interviews in a collection, searched together for a theme.
Questions
Can AI transcribe interviews accurately?
Yes, when the audio is clear. On the Audicite benchmark’s two-to-four-person dialogues the word error rate was under 1% and 95 to 99% of words were attributed to the right speaker, depending on noise.
Can AI tell different speakers apart?
Yes. Audicite labels each speaker automatically and refines the labels on the full recording; you can correct any line.