Transcribe a long Sony FX3 recording from the audio, not the master file.
A 5-hour Sony FX3 shoot day comes back as text in about 10 minutes when you upload its audio or a proxy rather than the XAVC master, which is far larger than the 5 GB a file may be. The first 20 minutes are free, and the 300 minutes fit the $6 pack of 500 minutes.
From the cinema camera to the text
An FX3 master is built for grading, not for sending. The transcript needs only the voices.
The XAVC folders
Sony cameras write XAVC video into the PRIVATE folder, then M4ROOT, then CLIP, with small XML files beside each clip. If proxy recording was switched on, a smaller copy of each clip sits on the card as well, usually in a SUB folder next to CLIP; check your own card.
Why the master is too big
Size is bitrate times duration. A clip recorded at 150 Mbps is about 67.5 GB an hour, and at 600 Mbps, about 270 GB. Even short masters pass 5 GB, so the master is never the file to upload here.
Send audio or a proxy
Export the interview's audio from your editor as WAV or MP3, or upload the proxy if it is under 5 GB. If you recorded sound on a separate recorder, that file is the cleanest choice. The words come out the same from any of them, with timestamps that match the shoot.
What the shoot becomes
A day of interviews returns with a brief per file, chapters with timestamps, key quotes and the names, numbers and dates said. The interviewer and subject are named in the summary, quotes and chapters, which is where paper edits usually begin.
Cinema camera files elsewhere
An FX3 master can be larger than whole transcription quotas.
✗ The usual walls
The better the codec, the further it is from an upload box.
- The OpenAI Whisper API stops at 25 MB
- Masters hundreds of gigabytes in size
- Free tiers that end before lunch break
- Logging selects by rewatching every clip
- Paper edits typed from memory
- Timecode lost in a copy-paste transcript
✓ Working from audio
A small file per interview, a full transcript back.
- WAV, MP3 or a proxy up to 5 GB
- A 5-hour day back in about 10 minutes
- Key quotes with timestamps for selects
- SRT captions to drop into the timeline
- 20 minutes free to check the sound
- A shoot day on the $6 pack
Sample
A sit-down documentary interview shot on a cinema camera, as it comes back.
This sample is illustrative, written for this page rather than drawn from any production. Roles stand in for names, and the timestamps show one interview in a 5-hour day.
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
Sony FX3 recordings, the common questions.
How long does a 5-hour FX3 shoot take to transcribe?
Close to 10 minutes, since each audio hour takes about 2. The public proof page records a 9.31-hour file done in 1,033 seconds.
What does a documentary's worth of rushes cost?
A single 5-hour day is 300 minutes on the $6 pack of 500 minutes. Thirty hours of interviews across a production come to 1,800 minutes, which fits the $12 pack of 2,000. Your first visit adds 20 free minutes.
Can I upload the XAVC master?
Only if it is 5 GB or less, and 2 GB or less before a pack. Most masters are far bigger, so export the audio or use the proxy; the transcript is the same.
Should I split a long take?
Only a take over 10 hours needs it. Upload each clip's audio as it is, or join a day's audio into one file if you want a single document.
Which languages work?
99, recognised automatically from the speech. An interview that moves between two languages is transcribed on its main one, and English with Tagalog in one file is handled.
What exports are there, and is the footage private?
Transcripts as TXT and DOCX, captions as SRT and VTT, data as JSON. Your files are transcribed on our own servers and never used to train any model; delete one and its transcript is deleted with it.
Related
Other cameras and long shoots
Send the shoot's audio. Read the first 20 minutes of the shoot free.
Transcribe my FX3 shootOperated by Neugence Technology Pvt. Ltd. · contact@whipscribe.com · Security · Privacy · Terms