pyannote-metrics
A toolkit for reproducible evaluation, diagnostic, and error analysis of speaker diarization systems
Decision gist · record as of 2026-08-14
Yes, if you are building or evaluating speaker diarization systems. The package is actively maintained, has low install friction, and is a standard tool in the diarization research community. However, verify the license terms first, as the metadata does not declare one explicitly. Not relevant for applications outside speaker diarization.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires Python >= 3.10.
- Low friction installation with a pure-Python wheel and six standard scientific dependencies (numpy, pandas, scikit-learn, scipy) plus two pyannote libraries.
- Actively maintained as of 100 days ago.
License · maintenance · safety
(unclear) — License status is unclear; no SPDX identifier or raw license text is available in the package metadata. Verify the actual license terms before use in proprietary or restricted contexts.
last release 2026-05-06 (100 days)
0 known vulnerabilities (OSV.dev, 2026-08-14) · 2,892,700 downloads/mo, #2,837 on PyPI
Alternatives
Verify before relying
pip install pyannote.metrics
from pyannote.metrics import DiarizationErrorRate
metric = DiarizationErrorRate()
der = metric(reference, hypothesis)- What license governs pyannote.metrics (unclear in metadata)?
- Does the package support evaluation of speaker overlap or only sequential speaker turns?
- What is the computational cost for large audio files or many speakers?
What it is and what it does
pyannote.metrics is a toolkit for evaluating speaker diarization systems—the task of determining who spoke when in an audio recording. It computes standard metrics (like diarization error rate), provides diagnostic breakdowns of errors, and supports reproducible evaluation workflows. The package wraps numpy, pandas, scikit-learn, and scipy to handle metric computation and analysis, and integrates with pyannote-core and pyannote-database for data representation.
It is designed for researchers and engineers building or benchmarking speaker diarization pipelines. Rather than a diarization system itself, it is the measurement and analysis layer: you provide reference (ground truth) and hypothesis (predicted) speaker segmentations, and the toolkit computes how well they match, where errors occur, and why. The package has been actively maintained since 2014 and is widely used in the speech processing community.
Use it for
- Compute diarization error rate (DER) and other standard metrics to benchmark a speaker diarization model against a reference annotation.
- Analyze where a diarization system fails—false alarms, missed speech, speaker confusion—to guide model improvements.
- Generate reproducible evaluation reports for research papers or system documentation.
- Compare multiple diarization hypotheses against the same reference to rank or select the best performer.
- Integrate metric computation into a diarization pipeline's validation or test harness.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you are building or evaluating speaker diarization systems.
The package is actively maintained, has low install friction, and is a standard tool in the diarization research community. However, verify the license terms first, as the metadata does not declare one explicitly. Not relevant for applications outside speaker diarization.
Install
pyannote-metrics on PyPI
Before you install
Low friction installation with a pure-Python wheel and six standard scientific dependencies (numpy, pandas, scikit-learn, scipy) plus two pyannote libraries. Actively maintained as of 100 days ago.
Requires Python >= 3.10.
License in practice
License status is unclear; no SPDX identifier or raw license text is available in the package metadata. Verify the actual license terms before use in proprietary or restricted contexts.
Quickstart
pip install pyannote.metrics
from pyannote.metrics import DiarizationErrorRate
metric = DiarizationErrorRate()
der = metric(reference, hypothesis)Verify before relying
- What license governs pyannote.metrics (unclear in metadata)?
- Does the package support evaluation of speaker overlap or only sequential speaker turns?
- What is the computational cost for large audio files or many speakers?
Package facts
| License | Not declared unclear |
| Python support | Supports the current Python release >=3.10 |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 6 packagesnumpypandaspyannote-corepyannote-databasescikit-learnscipy |
| Maintenance | Actively maintained 100 days since the last release |
| First released | |
| Downloads | 2,892,700 / month, #2,837 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
Evidence: pyannote_metrics-4.1-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “speaker diarization evaluation metrics”
- pyannote-metricsEvaluates and analyzes speaker diarization systems by computing…
- pyannoteai-sdkClient library for pyannoteAI's speaker diarization, speaker…
- pyannote-audioSpeaker diarization toolkit that identifies and separates individual…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also pyannote-audio · pyannote-core · pyannote-database · pyannoteai-sdk · Resemblyzer · texterrors · fairlearn · sherpa-onnx-core · speechmatics-voice · sherpa-onnx