Get hours of recordings back as searchable text
Send a folder of recordings (calls, depositions, interviews, meetings) and get transcripts back, usually within 48 hours. Your team stops typing and starts searching.
What you get back
- One transcript per file, in the format you choose (text, timestamps, or subtitles)
- A short report on what was run and how it performed
- Optional summaries or notes per call, scoped with you up front
A real run on public data
- Model: Whisper medium
- Input: 500 LibriSpeech test-clean files, 63.5 minutes of audio
- Word error rate: 2.29%
- Failures: 0
- Wall time: 29.9 minutes on CPU
Transcripts, script and evaluation: huggingface.co/datasets/booleai/librispeech-test-clean-whisper-medium-transcripts. LibriSpeech is clean read speech; accuracy on your recordings depends on audio quality. This is not a customer result.
Is it worth it? Do the math with your numbers
Compare a flat price with the staff hours it would take to transcribe by hand. Example only: if manual transcription took your team about 4 hours per audio hour (an assumption; it varies), about 50 audio hours would be about 200 staff hours. At the BLS OEWS, May 2025 national mean wage for paralegals, $33.51/hour, that is about $6,702 of staff time.
Separate transcription jobs were previously quoted from $1,000, agreed in writing before any work. The private AI sprint on this site is the flat $3,000 deployment below. There is no checkout link.
$3,000 flat
No ongoing fees. You own the deployment. Mutual NDA before kickoff.
Live within 48 hours of kickoff and data access, or a full refund.
And your files stay private
A private deployment lowers the risk of AI leaks. It doesn't replace your own security program and doesn't remove any compliance, ethics or contractual obligations (including SOC 2, if you're subject to it). We don't hold SOC 2 or any other certification. It can run fully offline on your hardware. Client data stays inside your environment and goes to no outside AI vendor. Full details: how we handle your data.