Python module for evaluating speaker diarization.
Project description
Getting Started
This project is Speechmatics' fork of https://github.com/pyannote/pyannote-metrics used to calculate various speaker diarization metrics from reference/hypothesis transcript pairs.
To install the package:
pip install speechmatics-diarization-metrics
This package has a CLI supporting ctm, lab, or V2 JSON format transcripts and can be run using:
python3 -m sm_diarization_metrics.cookbook reference.json transcript.json
For further guidance run:
python3 -m sm_diarization_metrics.cookbook -h
Run from source code
If you would prefer to clone the repo and run the source code:
git clone git@github.com:speechmatics/speechmatics-python.git
cd speechmatics-python/metrics/sm_diarization_metrics
pip install -r ./requirements.txt
python3 -m sm_diarization_metrics.cookbook reference.json transcript.json
Build wheel
To build and install the wheel run:
make wheel
make install
Docs
Further description of how to use the tool and the metrics available are in sm_diarization_metrics.pdf
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
Hashes for speechmatics_diarization_metrics-0.0.2-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 84ccba52f06e0105b5118b2b23a99e791810fd09e810fc1ef1056536389e0bc0 |
|
MD5 | 56329b88634ea1fd4c5638fb1d9be694 |
|
BLAKE2b-256 | 9aa53e01500412729fe545dc3e154678296094f460e64f34f73862940379da34 |