Speaker diarization · Word-level timecodes · Five export formats

Turn every recording into a transcript you can trust.

Verquil transcribes video and audio for research, legal, and medical teams — batch upload, speaker separation, word-level timecodes, and a custom vocabulary for your terms. 13 languages, exported as txt, docx, pdf, srt, or vtt. Punctuation is restored and checked word by word against the original, which is kept either way.

ORIGINAL
always kept
LANGUAGES
13
EXPORTS
5
Built for demanding teams
Research labsLaw firmsMedical centersAcademic journalsPublic sector
01 — Languages

13 languages, most detected for you.

Transcribe interviews, hearings, and clinical rounds in the language they were spoken — English, Traditional and Simplified Chinese, Japanese, Korean, Spanish and eight more. Auto-detect resolves 10 of the 13 without you picking anything, and handles code-switching within a single recording in one pass; for Chinese and Korean, select the language explicitly. Speaker labels stay consistent across the whole file.

auto-detect across 10 of 13
  • English
  • 繁體中文
  • 簡體中文
  • 日本語
  • 한국어
  • Español
  • Deutsch
  • Français
  • Português
  • Nederlands
  • Italiano
  • हिन्दी
  • Русский
02 — Accuracy

Accurate, and tunable to your field.

Speaker diarization and word-level timecodes are built in. Add your own keyterms so drug names, statutes, and proper nouns come back spelled the way you write them — not the way they happen to sound.

keyterm weighting · field-tuned
PembrolizumaboncologyCalibrated
Fed. R. Evid. 801(d)litigationCalibrated
Informed Consent v3.2clinicalCalibrated
Primary Endpoint (ORR)trialsCalibrated

Batch upload, billed per file

Drop in a batch of files at once. Every file transcribes independently and succeeds or fails on its own. One credit per minute of audio; failed jobs are refunded automatically, so you're never over-charged.

txt · docx · pdf · srt · vtt export

Every word checked against the original

Punctuation is restored where a recording comes back without it — most often in Chinese. Every restored passage is compared with the original word by word; if a passage does not match, you get the original text, not a guess. The original transcript is kept either way, so you can always show it exactly as it was produced. Most files also arrive with a short summary above the transcript; recordings with little speech, or where a summary cannot be produced, arrive without one.

03 — Pricing

Pay for minutes, not seats.

Verquil bills in credits — one credit per minute of audio, and failed jobs are refunded automatically. Published rates are on the way.

credit-based · per-minute
Coming soon

Published rates are being finalized

You don't have to wait to try it: every new account starts with 30 free minutes, no credit card required.

Start with 30 free minutes

Send your next batch to Verquil.

30 minutes free, no credit card. Batch upload, speaker separation, word-level timecodes, and five export formats. Punctuation is restored and verified word by word; the original transcript is always kept.