How It Works
FrameQuery transcribes your footage during indexing with full speech-to-text, word-level timestamps, and automatic speaker diarization. Find a quote from any interview, filter by speaker, and search by meaning with semantic embeddings - not just exact words.
We spent three weeks scouting locations before we found the right valley. The light had to be perfect for what we were trying to capture.
And the permits took even longer than that. You can't just show up with a crew and start filming on protected land.
Right. But once we had everything sorted, the first two days of shooting were incredible. Golden hour lasted almost forty minutes up there.
I remember the sound team had issues with the wind though. We ended up using the backup lavs for most of the exterior dialogue.
That's true. The boom was basically unusable above the treeline. Lesson learned for next time.
We should budget for a dedicated wind rig on the next shoot. It would have saved us two days of ADR.
Agreed. And the drone footage from day three more than made up for it. Those wide aerials are probably the best shots in the whole film.
The client loved those. That single tracking shot over the ridge ended up in every version of the cut.
Click any line to jump to that moment in the video
Search Syntax
"quarterly goals"Exact phrase match. Finds the precise moment someone said this.
-interviewExclude a term. Combine with other queries to narrow results.
@SarahFilter by speaker. Shows only transcript segments from that person.
codec:prores res:4kMetadata filters to narrow by format, resolution, camera, and more.
Semantic Search
MiniLM word embeddings run alongside keyword search on every query. Type what you mean and FrameQuery finds transcript segments that express the same idea, even when the words are completely different.
Searching by meaning, not just exact words. None of these results contain “discussing” or “budget” together.
We need to figure out the costs before we can commit to anything this quarter.
The finance team flagged our spend on the infrastructure side. It's higher than projected.
We talked about the budget allocation and they want a revised breakdown by Friday.
MiniLM word embeddings match the meaning of your query against transcript segments. “Discussing the budget” finds “figure out the costs” and “flagged our spend” because they mean the same thing, even though the words are completely different. Runs alongside keyword search on every query.
Navigate
Click any timestamp to jump directly to that moment in the video. The transcript highlights and scrolls in sync as playback continues, with speaker labels so you always know who is talking.
EXT_PARK_INTERVIEW.mov
32m 07s · 4.2 GB · 3840x2160
But once we had everything sorted, the first two days of shooting were incredible.
I remember the sound team had issues with the wind though. We ended up using the backup lavs.
The boom was basically unusable above the treeline. Lesson learned for next time.
We should budget for a dedicated wind rig on the next shoot.
Agreed. And the drone footage from day three more than made up for it.
Click any line to jump to that moment
Export
Select transcript segments and export them as timeline-ready files.
FAQ
FrameQuery uses Whisper for transcription, supporting 90+ languages with automatic language detection. You don't need to specify the language in advance.
Yes. Your transcript search runs across every indexed video in your library. Results are ranked by relevance and show which video and timecode each match is from.
Keyword search finds exact word matches using BM25 scoring. Semantic search uses MiniLM word embeddings to match by meaning, so “discussing finances” can find “we need to talk about the budget”. Both run simultaneously.
No. Transcription runs as part of the standard indexing pipeline on every plan, including Free.