Transcribe long recordings archived in Backblaze B2 one bucket, years of talks, finally searchable.
To transcribe a recording archived in Backblaze B2, download the object from the bucket through the web console or a command-line tool and upload the file here; a 7-hour recording is text in about 14 minutes. A single 420-minute file fits the $6 pack of 500 minutes, and an archive of forty 90-minute lectures, 3,600 minutes, fits the $24 pack of 5,000 minutes.
Cold archive to working text
B2 is a cheap place to keep recordings. It is a hard place to search them.
Getting the object out
For a handful of files, the bucket browser in the Backblaze web console lets you pick an object and download it. For dozens, a command-line tool such as the B2 CLI or rclone copies a whole folder to your disk in one go. Menu names change, so follow whatever your console shows for downloading a file.
Why the bucket URL is not enough
Even a public B2 file URL is not one of the link types we fetch; those are YouTube, Vimeo, Twitch, Kick, Google Drive and podcast feeds. Upload the downloaded copy instead. Files under 500 MB go into the box above, and larger ones continue on the upload page.
Pricing a whole archive
Forty lectures of 90 minutes each come to 3,600 minutes. The $24 pack of 5,000 minutes covers them and leaves 1,400 minutes, enough for another fifteen lectures of the same length. Credits never expire, so the archive can go up over weeks.
What each file returns
Every recording gets a brief with the thesis and takeaways, chapters with timestamps, the names, figures and dates mentioned, and the questions it answers. Keep the TXT or JSON next to the original object and the archive becomes something you can grep.
Why archives stay dark
Every recording is backed up. None of them can be searched.
✗ A bucket of media
Objects with dates for names and no idea what is inside.
- File names like 2019-03-14_final.mp4
- Hours of playback to find one talk
- Online tools that refuse anything large
- Splitting long files to fit upload caps
- A subscription for a one-time backlog
- No text to index next to the object
✓ A bucket with transcripts beside it
Download, upload, and store the text alongside the media.
- Seven hours processed in about 14 minutes
- Single objects up to 5 GB
- 3,600 minutes inside one $24 pack
- Chapters to tell talks apart
- JSON to feed your own index
- Packs that wait for the next batch
Sample
An archived seminar recording, pulled out of a B2 bucket.
Written for this page as an illustration; no real archive or speaker is quoted.
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
- AI chat over the whole 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
- Summary, chapters and key quotes
- All export formats
- AI chat over the whole transcript
Bulk pack
$24/5,000 min
83 hours of audio — $0.29 an hour. Credits never expire.
- 5,000 minutes of audio
- Summary, chapters and key quotes
- API access for batch pipelines
- All export formats
FAQ
Backblaze B2 archives, answered.
Can WhipScribe read straight from my bucket?
No. There is no connection to B2 and no access key to give us. Download the object yourself and upload it; that also keeps your bucket credentials out of anyone else's hands.
How quickly does a 7-hour object come back?
Around 14 minutes after the upload, since the pace is about 2 minutes per audio hour. For an archive, upload files in a steady stream and the transcripts collect in your Library.
How much does the whole archive cost?
Add up the minutes. Forty 90-minute lectures, 3,600 minutes, fit the $24 pack of 5,000 minutes. A smaller backlog of 1,500 minutes fits the $12 pack of 2,000 minutes, and one long file fits the $6 pack of 500 minutes.
Some objects are over 2 GB. Is that a problem?
Only before your first pack, when the limit is 2 GB. With a pack on the account, any file up to 5 GB or 10 hours is accepted. A recording longer than 10 hours goes up as two files.
Do I need to convert the files first?
Common audio and video files such as MP3, M4A, WAV and MP4 go up as they are. If an object is a format your browser will not offer in the file picker, convert it to MP3 or M4A first and check the result plays.
What happens to the copies I upload?
Each copy is transcribed on our own servers and is never used for training. The original stays in your bucket; delete the uploaded copy and transcript here whenever you like.
Related
More on archives and large files
Download the archive, upload it file by file. Seven hours, about fourteen minutes each.
Transcribe my B2 recordingsOperated by Neugence Technology Pvt. Ltd. · contact@whipscribe.com · Security · Privacy · Terms