Sample rate for transcription 16 kHz is enough; keep what the recorder made.
For transcription, 16 kHz is enough: it captures sound up to 8 kHz, which covers the range that makes speech intelligible, and files at 44.1 or 48 kHz work just as well. Change the rate only to shrink an uncompressed file, with ffmpeg -i in.wav -ar 16000 -ac 1 out.wav or your editor's export settings: a 48 kHz 24-bit stereo WAV is about 1.04 GB an hour and passes the 5 GB file limit at about 4.8 hours, while 16 kHz mono is about 115 MB an hour.
Sample rate, by the numbers
In a WAV, size is the sample rate times bit depth times channels.
The WAV table
Per hour: 16 kHz 16-bit mono is about 115 MB; 44.1 kHz 16-bit stereo about 635 MB; 48 kHz 16-bit stereo about 691 MB; 48 kHz 24-bit stereo about 1.04 GB; 96 kHz 24-bit stereo about 2.07 GB. The last one passes 2 GB in under an hour.
When the rate matters
Field recorders often default to 48 kHz 24-bit stereo. That file is over 2 GB a little before 2 hours and over 5 GB at about 4.8 hours. For an all-day session, set 44.1 or 48 kHz 16-bit mono, or record MP3, before you press record.
When it does not
In an MP3, M4A or AAC file the size follows the bitrate, not the sample rate, so a 44.1 kHz and a 48 kHz MP3 at 64 kbps are the same size. Those files go to the upload page as they are.
Phone and 8 kHz audio
Telephone recordings at 8 kHz carry speech up to about 4 kHz. They transcribe, but raising them to 44.1 kHz adds no detail, only bytes. Upload them at their original rate and expect the usual limits of a phone line.
Where sample-rate choices go wrong
Most problems are a studio setting left on for a five-hour talk.
✗ Studio settings on speech
High rates and bit depths built for music make long speech files huge.
- 96 kHz 24-bit for a lecture
- A day-long WAV past 5 GB
- Upsampling 8 kHz calls to 48 kHz
- Resampling twice through two apps
- Cutting the file to fit a size cap
- Expecting a higher rate to fix a far mic
✓ Speech settings, one file
Set the rate for voice, then let the file run the whole session.
- 16 or 44.1 kHz, 16-bit, mono
- 48 kHz is fine if the camera insists
- Keep phone audio at its own rate
- Up to 10 hours or 5 GB in one file
- About 2 minutes per audio hour
- $6 pack of 500 minutes covers 8 hours 20 minutes
Sample
A field interview recorded at 48 kHz, folded to 16 kHz mono, four hours in.
Illustrative only: these lines were written for this page. Resampling keeps the timeline, so the timestamps match the original WAV.
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
Sample rate, answered.
Is 16 kHz really enough for transcription?
For speech, yes. The sounds that tell words apart sit mostly below 8 kHz, which 16 kHz captures. Speech recognition commonly works on audio at around that rate, so a higher-rate file is not at a disadvantage and a 16 kHz file is not either.
Should I convert 48 kHz to 16 kHz before uploading?
Only to shrink a large WAV. ffmpeg -i in.wav -ar 16000 -ac 1 out.wav takes a 48 kHz 24-bit stereo hour from about 1.04 GB to about 115 MB. If the file is already under the limit, upload it as it is.
What sample rate should my recorder use for a long session?
44.1 or 48 kHz at 16-bit mono, or MP3 at 64 kbps if the recorder offers it. Ten hours of 48 kHz 16-bit mono WAV is about 3.46 GB, inside the 5 GB limit once a pack is on the account.
My file is over 2 GB. Can I still try it?
Before the first pack, uploads are limited to 2 GB, so resample or convert to MP3 first; a 64 kbps MP3 hour is about 29 MB. Once a pack is on the account, files up to 5 GB go through as they are.
How long and how much for a 5-hour WAV?
About 10 minutes once uploaded. Five hours is 300 minutes on the $6 pack of 500 minutes, with 200 left and no expiry. The first 20 minutes are free for a first-time visitor.
What comes back?
A brief with the thesis and takeaways, chapters with timestamps, the names, numbers and dates said, and a chat over the transcript; speakers are named in the summary, key quotes and chapters. Exports are TXT, DOCX, SRT, VTT and JSON. Recordings are transcribed on our own servers and never used for training.
Related
Related pages
Keep the rate your recorder chose. Upload, and read it in minutes.
Transcribe my recordingOperated by Neugence Technology Pvt. Ltd. · contact@whipscribe.com · Security · Privacy · Terms