# TimoBolkart/voca

This codebase demonstrates how to synthesize realistic 3D character animations given an arbitrary speech signal and a static character mesh.

Repository: https://github.com/TimoBolkart/voca
Canonical: https://ross.abutalabs.com/products/timobolkart-voca
Homepage: https://voca.is.tue.mpg.de/en
Language: Python
License Family: other
Topics: computer-vision, computer-graphics, face-animation, 3d-models, morphable-model, machine-learning, 3d-face, python, tensorflow, voca, animation-sequence, python3
Last push: 2024-08-20T10:53:36+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 2694, "days_push": 743, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1263, forks 277 (observed 2026-08-28T04:04:10.448103+00:00)

## What it is
VOCA is a speech-driven 3D facial animation framework that synthesizes realistic character face animations from an arbitrary speech signal and a static character mesh. It is the official research code accompanying the CVPR 2019 paper 'Capture, Learning, and Synthesis of 3D Speaking Styles', built on Python and TensorFlow.

## Use cases
- animate a 3D character mesh from an audio speech clip
- generate talking-head facial animation from voice recordings
- research speech-driven 3D face animation
- apply one trained model to animate multiple character identities
- produce 3D speaking-style animation sequences for games or films
- work with the FLAME morphable face model

## When to choose
- you need research-grade speech-to-3D-face animation with a pretrained model
- you work with FLAME topology meshes and TensorFlow 1.x
- you want to reproduce or build on the VOCA/VOCASET CVPR 2019 work

## When to avoid
- you need production-ready software with an active support community
- you cannot use legacy TensorFlow 1.x and Python 3.6/3.7 environments
- you need real-time lip sync in a game engine rather than offline mesh animation
- you require a permissive license - the repo has no license file

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, audio-processing, image-processing, animation
- domain: computer-vision, machine-learning, deep-learning
- platform: python
- tags: speech-driven-animation, 3d-face, facial-animation, morphable-model, flame, tensorflow, research-code, talking-heads, computer-graphics, linux, gpu

## Member repositories
- TimoBolkart/voca (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:10.448103+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 2026-08-30T05:04:06.291229+00:00, confidence not recorded.
  - readme: https://github.com/TimoBolkart/voca (fetched 2026-08-28T04:04:10.448103+00:00, sha acf4135bb359)
- Data as of 2026-08-30T08:39:29.467469+00:00.