Transcribe a long ProRes MOV by sending the sound, not the master.
A ProRes MOV is edit-grade video: ProRes 422 at 1080p runs about 147 Mbps, roughly 66 GB an hour, so even one hour is far over the 5 GB file limit. Export the audio (about 58 MB an hour as a 128 kbps MP3) or a small proxy, upload that, and a 3-hour interview transcribes in about 6 minutes with the first 20 minutes free.
ProRes, worked out
An editing codec keeps every frame. The transcript needs none of them.
Why ProRes is so large
ProRes is an intermediate codec built for editing, so it keeps high bitrates instead of compressing hard. Apple's published target for ProRes 422 at 1080p and 29.97 frames is about 147 Mbps; that is 147,000,000 ÷ 8 × 3,600, around 66 GB an hour. Higher flavours and 4K are larger still.
The two lines, for a master file
A ProRes 422 master crosses 5 GB in under five minutes of footage, so the 10-hour line never comes into it. ProRes Proxy, the smallest flavour at roughly 45 Mbps for 1080p, still passes 5 GB inside 15 minutes.
Export the audio track
The sound in a ProRes MOV is often uncompressed PCM, around 1 GB an hour at 48 kHz 24-bit stereo. Export it from the editor as an MP3 or WAV, or run ffmpeg -i master.mov -vn -c:a libmp3lame -b:a 128k audio.mp3, and drop the result on the upload page.
Back into the edit
The transcript keeps the audio's own timeline, so if the export starts at the first frame, every timestamp lines up with the source. SRT and VTT bring subtitles back into the editor, and the chapters work as a first paper edit.
Why editors give up on transcribing masters
Sending a 66 GB master to read the dialogue does not add up.
✗ Uploading the master
An hour of ProRes 422 is about 2,600 times the OpenAI Whisper API's documented 25 MB upload limit.
- Overnight uploads of picture data
- Masters copied off the edit drive
- A clip cut into dozens of pieces
- Logging interviews by hand
- Timecode that no longer matches
- Nothing to search across the shoot
✓ Sound out, text back
An audio export of a few hundred megabytes carries every word the master does.
- About 58 MB an hour as 128 kbps MP3
- Audio exports up to 10 h or 5 GB
- About 2 minutes for each hour of dialogue
- Chapters for a paper edit
- SRT and VTT to drop on the edit
- $12 covers 2,000 minutes of rushes
Sample
A documentary interview, the second hour of the take.
This sample is illustrative, written for this page rather than drawn from a real shoot. The timestamps follow the exported audio, which starts with the master.
Export
One transcript. Five clean formats.
Every pack exports all five — no format is held back for a higher tier.
Plain text
De-ummed paragraphs. Ready to paste.
SRT captions
Timed to the speech. Every video editor reads this.
WebVTT
HTML5 player + YouTube uploads.
Show notes
Formatted with chapters and pull-quotes.
Machine-readable
Timestamped text for your own tools.
Pricing
Honest pricing, no surprises.
A pack from $6 (500 minutes) opens the full transcript. After that, no subscription and no auto-renew — every tier is a one-time minute pack, and credits never expire. Buy once, use the minutes whenever you need them.
Casual pack
$6/500 min
8 hours of audio — $0.72 an hour. One-time — credits never expire.
- 500 minutes of audio
- All export formats
- Unlimited AI chat per transcript
- Executive summary & key quote
Best value
$12/2,000 min
33 hours of audio — $0.36 an hour. Credits never expire.
- 2,000 minutes of audio
- Priority queue
- All export formats
- Unlimited AI chat per transcript
Bulk pack
$24/5,000 min
83 hours of audio — $0.29 an hour. Credits never expire.
- 5,000 minutes of audio
- Priority queue
- API access for batch pipelines
- All export formats
FAQ
Long ProRes MOV files, answered.
How long does a ProRes interview take to transcribe?
Once the audio is up, about 2 minutes per hour of recording: a 3-hour interview in roughly 6 minutes. Exporting the audio from the editor usually takes less time than uploading even a few minutes of the master would.
What does a shoot's worth of interviews cost?
Fifteen hours of rushes is 900 minutes, which fits the $12 pack of 2,000 minutes and leaves 1,100. A single 3-hour interview, 180 minutes, fits the $6 pack of 500 minutes. Anyone new gets the first 20 minutes of the interview free.
How big is ProRes per hour?
By Apple's published 1080p rates at 29.97 fps: Proxy about 45 Mbps (20 GB an hour), 422 LT about 102 Mbps (46 GB), 422 about 147 Mbps (66 GB), 422 HQ about 220 Mbps (99 GB). 4K is roughly four times those.
Can I upload a proxy instead of the audio?
Yes, if it fits: an H.264 proxy at 720p and around 2.5 Mbps is about 1.1 GB an hour. Audio alone is smaller and gives the same transcript. If the upload refuses the proxy, convert it to MP3 first.
Can the interviews be in other languages?
Yes. The language is detected automatically from 99, so a Japanese or Swedish interview needs no setting. Music beds and effects in a cut are not transcribed; the dialogue is.
Which exports, and what happens to unreleased footage?
TXT, DOCX, SRT, VTT and JSON, with speakers named in the summary, quotes and chapters. Unreleased audio is transcribed on our own servers and never used for training, and you can delete it after export.
Related
Related pages
Export the dialogue track and upload it. No charge for the first 20 minutes of dialogue.
Transcribe my interview audioOperated by Neugence Technology Pvt. Ltd. · contact@whipscribe.com · Security · Privacy · Terms