# resemble-ai/Resemblyzer

A python package to analyze and compare voices with deep learning

Repository: https://github.com/resemble-ai/Resemblyzer
Canonical: https://ross.abutalabs.com/products/resemblyzer
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2023-10-12T09:35:22+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 2575, "days_push": 1056, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3300, forks 479 (observed 2026-08-28T04:07:55.689151+00:00)

## What it is
Resemblyzer is a Python package that uses a deep learning voice encoder to convert speech audio into 256-dimensional voice embeddings. These embeddings enable voice comparison, speaker verification, diarization, and fake speech detection.

## Use cases
- compare how similar two voices sound
- verify a speaker's identity from a few seconds of reference audio
- diarize a multi-speaker recording to see who is talking when
- detect deepfake or fake speech
- extract voice embeddings for downstream ML models
- cluster utterances by speaker

## When to choose
- you need speaker embeddings or voice similarity in Python with minimal setup
- you want speaker diarization or verification without training a model
- you need a lightweight pretrained voice encoder for research or prototyping

## When to avoid
- you need full speech-to-text transcription
- you need production-grade text-to-speech or voice cloning
- you require actively maintained software with recent releases
- you need real-time streaming speaker recognition

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, audio-processing, speech-recognition, nlp
- domain: machine-learning, speech-processing, artificial-intelligence
- platform: python, cross-platform
- tags: voice-encoder, speaker-verification, speaker-diarization, voice-embeddings, deep-learning, audio-embeddings, audio

## Member repositories
- resemble-ai/Resemblyzer (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:55.689151+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T18:42:57.157584+00:00, confidence not recorded.
  - readme: https://github.com/resemble-ai/Resemblyzer (fetched 2026-08-28T04:07:55.689151+00:00, sha bc137e885bfd)
  - registry_pypi: https://pypi.org/pypi/resemblyzer/json (fetched 2026-08-29T09:35:30.744403+00:00, sha 523d2ba83be5)
- Data as of 2026-08-30T08:39:29.467469+00:00.