# AI use cases for Pyannote Audio (pyannote/pyannote-audio)

Repo: [pyannote/pyannote-audio](https://github.com/pyannote/pyannote-audio) | Use cases: 3 | Page: https://dedupedai.com/use-cases/pyannote-pyannote-audio/

pyannote.audio is an open-source Python toolkit for speaker diarization built on PyTorch. Its community-1 pipeline runs locally from Hugging Face and tells you who spoke when, and the same API can call the premium precision-2 pipeline on pyannoteAI servers. Models can be fine-tuned on your own data, and optional telemetry can be turned off.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Find who spoke when, locally: https://dedupedai.com/use-cases/pyannote-pyannote-audio/find-who-spoke-when-locally-with-pyannote/
- Try the premium precision-2 pipeline: https://dedupedai.com/use-cases/pyannote-pyannote-audio/use-precision-2-premium-diarization-with-pyannote/
- Turn off telemetry for private diarization: https://dedupedai.com/use-cases/pyannote-pyannote-audio/turn-off-pyannote-telemetry-for-private-diarization/

## How to find who spoke when, locally with Pyannote Audio

Repo: [pyannote/pyannote-audio](https://github.com/pyannote/pyannote-audio) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Run the community-1 open-source diarization pipeline on your own machine and print each speaker turn with start and stop times.

**Who it is for:** A developer adding speaker labels to transcripts.

**The problem:** Transcripts without speaker turns are hard to read and analyze.

**The outcome:** A list of turns like start=0.2s stop=1.5s speaker_0.

### Steps

1. Install ffmpeg, then `pip install pyannote.audio`
2. Accept the pyannote/speaker-diarization-community-1 user conditions and create a Hugging Face access token
3. Load the pipeline with Pipeline.from_pretrained("pyannote/speaker-diarization-community-1", token=...) as in the README
4. Send it to the GPU when available, run it on audio.wav with a ProgressHook, and loop over output.speaker_diarization to print turns

### Prompt to paste into your agent

```text
Using pyannote/pyannote-audio, write a script that loads speaker-diarization-community-1 with my HF token from an environment variable, runs it on <my audio file> with ProgressHook, and saves the turns to a CSV.
```

### Caveat

Speaker labels are anonymous per file; mapping them to real names is up to you.

Risk, credentials: Keep your Hugging Face token in an environment variable rather than hard coded in scripts.

Risk, privacy: Speaker analysis of recordings of other people needs their consent.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/pyannote-pyannote-audio/find-who-spoke-when-locally-with-pyannote/

## How to try the premium precision-2 pipeline with Pyannote Audio

Repo: [pyannote/pyannote-audio](https://github.com/pyannote/pyannote-audio) | Difficulty: Intermediate | Time: 30 minutes | Basis: From the README

Swap the pipeline name and token to run diarization on pyannoteAI servers with the same Python API.

**Who it is for:** A team comparing open-source quality against a hosted service.

**The problem:** Hard audio with many speakers can need a stronger model than you can host.

**The outcome:** Speaker turns from the precision-2 service, printed the same way as the local pipeline.

### Steps

1. Install with `pip install pyannote.audio`
2. Create a pyannoteAI API key at dashboard.pyannote.ai, which comes with free credits
3. Load Pipeline.from_pretrained("pyannote/speaker-diarization-precision-2", token=...) with that key
4. Run it on audio.wav and loop over output.speaker_diarization, then compare the turns with your community-1 run

### Prompt to paste into your agent

```text
With pyannote/pyannote-audio, run both community-1 locally and precision-2 with my PYANNOTEAI_API_KEY on <one test recording>, and show me where the speaker turns differ.
```

### Caveat

Speaker names come back as SPEAKER_00 style labels from the premium service.

Risk, privacy: precision-2 runs on pyannoteAI servers, so your audio leaves your machine.

Risk, cost: Free credits run out; check pyannoteAI pricing before processing a large archive.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/pyannote-pyannote-audio/use-precision-2-premium-diarization-with-pyannote/

## How to turn off telemetry for private diarization with Pyannote Audio

Repo: [pyannote/pyannote-audio](https://github.com/pyannote/pyannote-audio) | Difficulty: Beginner | Time: 15 minutes | Basis: From the README

Disable pyannote's optional usage metrics for a session or globally before processing sensitive audio.

**Who it is for:** Teams handling confidential recordings, such as legal or medical audio.

**The problem:** Compliance rules may forbid any outbound usage reporting from processing machines.

**The outcome:** Diarization that runs locally with metrics disabled.

### Steps

1. Install with `pip install pyannote.audio`
2. Disable metrics in your shell with `export PYANNOTE_METRICS_ENABLED=0`
3. Or in Python call set_telemetry_metrics(False, save_choice_as_default=True) from pyannote.audio.telemetry to make it the default
4. Run the local community-1 pipeline as usual

### Prompt to paste into your agent

```text
Using pyannote/pyannote-audio, set PYANNOTE_METRICS_ENABLED=0, save the telemetry choice globally with set_telemetry_metrics, then run community-1 on <my audio> and confirm the setting.
```

### Caveat

The README says telemetry tracks file duration and speaker count settings, not who the user is; the first model download still contacts Hugging Face.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/pyannote-pyannote-audio/turn-off-pyannote-telemetry-for-private-diarization/
