Problem[01]
A collection nobody can search
Hundreds of hours of filmed interviews, and the only finding aid is a summary sheet per session. The detail of what narrators said is out of reach.
For Oral history and research interviews
Every filmed testimony, transcribed and searchable
Two-hour life-story interviews, recorded on video, some in a second language. Prewatch transcribes them on your own Mac with a time on every line, so the whole collection answers a search.
14 days free, then $59 a year or $7.99 a month. Works with your Claude or ChatGPT plan; timelines need DaVinci Resolve Studio 21.1+. Clip descriptions are optional and use your own API key, billed by Anthropic or OpenAI.
Works withA Mac with Apple silicon · macOS 26+
ClaudeClaude Desktop and Claude Code
ChatGPTThe ChatGPT Mac app and Codex
DaVinci Resolve Studio 21.1Timelines through its own MCP server
Any drive or NASThe Prewatch folder travels with the media
Problems
What slows Oral history and research interviews down
Problem[02]
A day of typing per hour
Transcribing an hour of testimony properly takes most of a working day. With no budget for it, the backlog grows faster than volunteers can work through it.
Problem[03]
The release form sets the limits
Narrators signed a release for your archive, not for a transcription service's servers. For many projects and ethics reviews, uploading the recordings is not an option.
What Prewatch gives you
Your footage, already watched
Prewatch indexes every clip on your Mac, so Claude or ChatGPT knows what is in your footage and can build the cut with you in DaVinci Resolve Studio.
| What you get | Where it runs | What it does for you |
|---|---|---|
| Transcription on the Mac | On your Mac | Whisper on the Mac's GPU, in the language spoken, every word timed. No audio or video leaves the Mac to be transcribed. |
| Interviewer and narrator apart | On your Mac | Speaker turns within each recording, so questions read apart from answers, and the same narrator recognized across recordings. Names are yours to give; voiceprints never leave your Mac. |
| Doubts marked, not hidden | On your Mac | Lines the recognizer was unsure of are marked unsure. A recording it could not hear credibly gets no transcript, with the reason noted. |
| Translation for researchers | Claude or OpenAI | Interviews in another language translated to English line by line with your own Claude or OpenAI API key. Without a key, Whisper translates on the Mac, rougher. |
| Plain, durable files | On your Mac | Transcripts as JSON, WebVTT and SubRip beside the recordings, plus one Markdown file of every line, readable without Prewatch. |
| One search across sessions | On your Mac | Search every indexed recording at once for a place, a family name or a phrase, and get the recording and the second. |
Prewatch · ClipsMac app
The screenshot arrives with the app
Ask it
One question instead of a day of scrubbing
You ask in Claude or ChatGPT. Prewatch answers with the clips, the seconds and the timecode, and your assistant builds the timeline in DaVinci Resolve Studio.
01
Where do narrators talk about the mill closing?
Every recording that mentions it, with the lines as transcribed and the time each passage starts.
Your AI app02
Outline what this interview covers, in order.
A topic outline drawn from the transcript, each topic with its start time, ready for a finding aid.
Your AI app03
Which interviews need a human check before we share the transcripts?
Recordings with no transcript and why, and every line marked unsure with its time, so reviewers listen there first.
Your AI appIs it accurate enough for an archive?
Treat it as a strong first draft, not a verified transcript. Lines Whisper was unsure of are marked and every line keeps its time, so a reviewer checks those against the recording before anything is published or cited.
Does it work with audio-only recordings?
No. Prewatch indexes video files that macOS can decode, in MOV and MP4. WAV, MP3 and other audio-only files are not indexed.
Does anything leave the Mac?
Not to transcribe or tell speakers apart. Voiceprints stay on your Mac and there is no face recognition. Descriptions and translation by a model are opt-in with your own key and send keyframes and transcript text, never audio, video or speaker labels. What you ask Claude or ChatGPT about a recording becomes part of that conversation.
Does it work with the free version of DaVinci Resolve?
Building timelines by chat needs DaVinci Resolve Studio 21.1 or later, whose built-in MCP server lets your assistant work in Resolve. With the free version you can still index your footage, search it and ask Claude or ChatGPT about it, then place the clips yourself. A file handoff that the free version can import is planned.
How does the free trial work?
Prewatch is an App Store subscription with a 14-day free trial: $59 a year or $7.99 a month after it. You manage or cancel it in your Apple Account's subscription settings; cancel before the trial ends and you are not charged.
Try it on your own footage
Point Prewatch at a folder you have already shot, then ask about it in Claude or ChatGPT.
14 days free, then $59 a year or $7.99 a month. Works with your Claude or ChatGPT plan; timelines need DaVinci Resolve Studio 21.1+. Clip descriptions are optional and use your own API key, billed by Anthropic or OpenAI.