Run pyannote speaker diarization
Ready in about 90 seconds. No install, no setup. From $0.22 per hour.
What it does
This model listens to a recording with several people in it and marks which stretches belong to which speaker.
It does not know their names. It gives you Speaker 1, Speaker 2, and so on, with timestamps, and you attach the names afterwards.
On its own it is not very useful. Run it alongside Whisper and you get a transcript where every line is attributed, which is what most people actually want from a meeting or interview recording.
What you can make
These are prompts, not fixed presets. The workspace opens with one of them already loaded so your first result is one click away.
Your setup
We run this on an RTX 3090. It processes about an hour of audio in 90 seconds, and it runs alongside Whisper on the same machine. $0.22 per hour, billed by the minute.
Why this card?
The model is small, under 100MB, so it barely touches memory. There is no reason to pay for a bigger card. The one thing worth knowing is that we run it on the same machine as the transcription rather than a separate one, because moving the audio twice costs more time than the diarization itself.
- Memory
- 24GB
- Speed
- about 2 minutes per hour of audio
- Rate
- $0.22/hr
What will it cost you?
Billed by the minute on a RTX 3090 at $0.22 per hour. A batch job shuts the GPU off when the last hour of audio is done, so this is the whole cost.
Two ways to run it
Click and use it
Upload a recording and see a timeline coloured by speaker.
Get early access Best for exploring.Hand us the whole pile
Usually bundled with transcription rather than run alone.
Join the list Best for volume. The GPU shuts off automatically when it is done.Model facts
| Parameters | 25M |
|---|---|
| License | MIT |
| Memory required | 4GB minimum |
| Base model | Trained from scratch |
| Publisher | pyannote |
| Recommended hardware | RTX 3090, 24GB |
| Hugging Face | pyannote/speaker-diarization-3.1 |
Want pyannote speaker diarization the day it opens?
We are testing the rental flow with a small group first. Leave an email and we will tell you when pyannote speaker diarization is ready to run. One message, no newsletter.
pyannote speaker diarization questions
Do I need to install anything to use pyannote speaker diarization?
No. We start a machine with pyannote speaker diarization already loaded and hand you a link. Everything runs in your browser, there is nothing to download, and nothing is left on your computer afterwards. It works the same on a Mac, a Windows laptop, or a Chromebook. A workspace is usually ready in about 90 seconds.
How much does it cost to run 40 hours of audio?
About $0.22. pyannote speaker diarization takes about 2 minutes per hour of audio on the RTX 3090 we recommend, so 40 hours of audio is roughly 60 minutes of GPU time at $0.22 per hour. Billing is by the minute, and a batch job shuts the GPU off the moment the last item finishes. The calculator above works this out for your own numbers.
Can I run pyannote speaker diarization on a cheaper card?
Sometimes, and the calculator will not always make it look worth it. The model is small, under 100MB, so it barely touches memory. There is no reason to pay for a bigger card. The one thing worth knowing is that we run it on the same machine as the transcription rather than a separate one, because moving the audio twice costs more time than the diarization itself. If you want to try a different card anyway, the advanced catalog lets you pick one and shows the estimated time before you commit.
Can I use pyannote speaker diarization commercially?
Yes. pyannote speaker diarization is released under the MIT license, which permits commercial use with no conditions worth worrying about. You can use the output in paid client work and in products you sell.
What happens to my files after I stop?
Anything you save into your workspace folder stays in your account and is there when you come back. Everything else is destroyed: the machine is terminated and its disk is wiped. Batch job inputs and outputs are deleted from our storage 24 hours after you download them. We do not read your files or use them to train anything.
What if it fails partway through?
You are not charged for work that did not complete, and failed sessions are refunded automatically. Batch jobs write each output as it finishes rather than at the end, so a failure at item 380 of 400 leaves you 379 usable files and a message saying what broke. Restarting picks up where it stopped rather than redoing the work.
Own a GPU that sits idle? Put it to work. The average provider earns $180 to $420 per month per card.
List your GPU