Measured on production · 3,006 jobs · 90 days to 22 Sep 2026

How fast is Whipscribe? A one-hour file in about two minutes.

Most transcription services put your file in a queue and email you later, with turnaround quoted in hours or days. Whipscribe runs on our own servers and starts the moment your upload lands. These are our real production numbers for the last 90 days: every completed job, every customer, failed jobs excluded.

First transcript free, second $0.99 · then from $0.36/hour · Credits never expire · Never used to train AI
Median 53 s upload to transcript · 87% start within 5 s · 21x real time · Never used to train AI

The numbers

What “fast” means here, with the measurement behind each claim.

53 seconds, median

From the moment the upload finishes to the transcript being ready, across 3,006 completed jobs in the 90 days to 22 September 2026 (2,054 hours of audio). Half of all jobs were done inside a minute, 70% inside two minutes, 86% inside five.

A one-hour file in 2 min 5 s

227 recordings between 55 and 65 minutes long. Median 2 min 5 s, and nine in ten were ready within 3 min 13 s. Files of 30 to 60 minutes: median 97 seconds, and that figure held at 106 s in July, 91 s in August and 103 s in September.

No queue to speak of

The median wait between upload and the start of processing is 0.4 seconds; 87% of jobs start within five seconds. There is no batch that runs at night and no email that arrives tomorrow. Very long files (over two hours) are the exception: median 20 minutes end to end.

21x real time, then straight to the screen

Processing runs at a median 21.6 times real time: 2.8 seconds of work per minute of audio. When it finishes, the transcript page paints in about 0.4 seconds. You read while the tab is still open, not from a link in your inbox.

Why it matters

A queue you wait in vs a transcript you read now.

✗ The queue model

Many services accept your file, place it in a processing queue and notify you later. Turnaround is quoted in hours, sometimes a business day, and the wait grows when the service is busy.

  • Turnaround quoted in hours or days
  • You leave and come back
  • The wait grows with demand
  • Speed is a premium tier
  • Result arrives by email

✓ Whipscribe, measured

The file starts processing the moment it lands and the transcript appears in the same tab. Speed is the same for the free first transcript as for a plan.

  • Median 53 s end to end
  • Processing starts in under a second
  • A one-hour file in about two minutes
  • Same speed at every price
  • Transcript on screen, exports ready

By file length

Upload to transcript, by how long the recording is.

Production data, 90 days to 22 Sep 2026, completed jobs only. Median is the typical case; p90 means nine in ten jobs finished within that time. Measured from the end of the upload; your own upload time depends on your connection (our uploads averaged 58 MB in 28 seconds).

Recording lengthJobs measuredMedian9 in 10 withinSpeed vs real time
Under 5 minutes1,1387 s38 s7.6x
5 to 30 minutes71145 s3 min 10 s19.8x
30 to 60 minutes4861 min 37 s3 min 8 s24.8x
1 to 2 hours3872 min 40 s6 min 23 s25.2x
Over 2 hours28520 min 16 s25 min 49 s32.0x

Export

One transcript. Five clean formats.

Every pack exports all five — no format is held back for a higher tier.

.txt

Plain text

De-ummed paragraphs. Ready to paste.

.srt

SRT captions

Word-level. Every video editor reads this.

.vtt

WebVTT

HTML5 player + YouTube uploads.

.docx

Show notes

Formatted with chapters and pull-quotes.

.json

Machine-readable

Per-word timing + speaker IDs.

Pricing

Honest pricing, no surprises.

Your first transcript is free at any length and your second is $0.99. 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

$8/1,000 min

16 hours of audio — $0.48 an hour. One-time — credits never expire.

  • 1,000 minutes of audio
  • All export formats
  • Unlimited AI chat per transcript
  • Executive summary & key quote
Get 1,000 minutes

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
Get 2,000 minutes

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
Get 5,000 minutes

FAQ

Questions about speed, answered from the data.

How were these numbers measured?

From our production job records: every job that completed in the 90 days to 22 September 2026, for every customer, with failed jobs excluded and our own test accounts removed. The clock starts when the upload finishes and the job is created, and stops when the transcript is ready. Medians and 90th percentiles are computed per file-length band. We will refresh this page as the data moves.

Does the time include my upload?

No. Upload time depends on your connection and file size; across our uploads it averaged 58 MB in 28 seconds. The turnaround figures on this page start the moment the upload has landed. Pasting a link instead of uploading a file adds the time to fetch the media: for 5 to 60 minute files the median was 86 seconds by link versus 56 seconds by upload.

Why are files over two hours slower?

Very long recordings are processed in larger units and can wait behind each other, so their median end-to-end time was 20 minutes in this period even though processing itself ran at 32 times real time. If you need a three-hour recording quickly, splitting it into hour-long files gets each part back in a few minutes.

Is speed different on the free first transcript?

No. Every job goes through the same path at the same priority. The figures above include first-time visitors, pay-as-you-go packs and plan customers alike.

Is it as fast in languages other than English?

The measurements cover every language our customers used in the period, auto-detected. We have not seen a language-specific slowdown in the data, but we publish the overall figures rather than per-language ones until each language has enough jobs to report honestly.

What happens on a bad day?

Sometimes a job waits longer, which is why we publish 90th percentiles beside medians: nine in ten one-hour files were ready within 3 min 13 s. If a job fails, you are told on the page and nothing is charged for it.

Related

Related pages.

Time it yourself. Upload a file now.

Try Whipscribe

Operated by Neugence Technology Pvt. Ltd. · contact@neugence.ai · Security · Privacy · Terms