Transcribe a long GSM file phone audio, read back.
A long GSM file holds phone-quality audio at 13 kbps, about 5.9 MB per hour, so a 10-hour recording is under 60 MB and length is the only limit, at 10 hours per file. It transcribes at about 2 minutes per hour of audio; the first 20 minutes are free and an 8-hour shift recording fits the $6 pack of 500 minutes.
GSM, worked out
Thirteen kilobits a second is enough for a voice. That makes the files almost weightless.
Where GSM files come from
GSM 06.10 is the full-rate codec from early mobile networks, and telephone systems adopted it for storing calls because it is small and cheap to encode. Office phone systems, call recorders and voicemail servers write it as raw .gsm files or as WAV files with GSM audio inside.
The size, worked out
At 13 kbps an hour is 13,000 bits a second times 3,600 seconds, divided by 8: about 5.85 MB. Ten hours is under 60 MB, a tiny fraction of the 2 GB that uploads before a pack. The 10-hour line per file is the only one a GSM archive can meet.
Getting raw GSM in
Upload the file on the upload page. A raw .gsm file has no header to describe it, so some tools cannot guess what it is; if the upload refuses the file, convert it to MP3 first. ffmpeg or sox can read 8 kHz GSM and write an MP3 in one command.
Calls you can search
A long call log comes back with a brief, chapters at each new call or topic, the account numbers, dates and names said aloud, and a chat that finds the moment a promise was made. The agent and the caller are identified and named in the summary, key quotes and chapters.
Why call archives go unread
A phone system can store years of calls. Nobody can listen to them.
✗ Calls stuck in a phone format
Raw GSM opens in few players, and narrowband audio is hard work for tools tuned to podcasts.
- Files with no header to identify them
- Players that will not open .gsm
- 8 kHz audio that other tools garble
- One file per call, and thousands of them
- Disputes settled by memory
- Quality reviews done by sampling
✓ A call log, in writing
Join a shift's calls into one file, or upload a long one as it is.
- Up to 10 hours per GSM file
- Processing takes about 2 minutes an hour
- The first 20 minutes come free
- $24 for 5,000 minutes, with API access
- Each call opens its own chapter
- Plain text, Word and JSON
Sample
A support line, five hours into a shift recording.
Written for this page as an illustration, this support call is not from any real customer. The timestamps show where each exchange falls in a long GSM file.
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
- Unlimited AI chat per 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
- Priority queue
- All export formats
- Unlimited AI chat per transcript
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
FAQ
Long GSM files, the common questions.
How long does a long GSM file take to transcribe?
Near 2 minutes per audio hour, so an 8-hour shift recording is ready in about 16 minutes. At under 60 MB, even a 10-hour file uploads in moments.
What does a call archive cost?
An 8-hour shift is 480 minutes, which fits the $6 pack of 500 with 20 to spare. Eighty hours of recorded calls is 4,800 minutes, on the $24 pack of 5,000. A support team that records every week can also use the $19-a-month Workspace.
How big is a GSM file per hour?
About 5.85 MB at the full-rate 13 kbps. A WAV file with GSM 6.10 audio inside carries a small header on top and is almost the same size. Neither comes near 2 GB, let alone the 5 GB ceiling.
Should I upload each call separately or join them?
Either works. Separate files give one transcript per call. Joining a shift into one file of 10 hours or less gives a single transcript with a chapter for each call, which is faster to search; ffmpeg can concatenate them in order.
Which languages are covered?
99, detected automatically, so a line answered in Spanish, Portuguese or Afrikaans needs no setting. Calls that switch between English and Tagalog are handled. Hold music is not transcribed; the conversation is.
What do I download, and where are calls processed?
Call transcripts download as TXT, DOCX, SRT, VTT and JSON. Calls are transcribed on our own servers and are never used for training, and you can remove any call recording and its text from the account yourself.
Related
Other phone and voice formats
Upload the GSM file. Twenty minutes of calls, free.
Transcribe my GSM fileOperated by Neugence Technology Pvt. Ltd. · contact@whipscribe.com · Security · Privacy · Terms