PiperHub – Piper TTS voices catalogue

piperhub.org is a free, open-source catalogue of voices for Piper TTS. Browse 490+ models and 3,600+ speakers in 70 languages, listen to samples, check where each voice comes from (model ↔ dataset), and synthesize speech locally in your browser.

piperhub.org est un catalogue libre et open source de voix pour Piper TTS : parcourez les modèles, écoutez des extraits, identifiez la source de chaque voix et générez de la parole localement dans votre navigateur.

• 🎧 Voices / Voix
• 📦 Datasets
• 📖 Guides · Piper Plus
• 🌍 16 langues : English (racine) + fr, de, es, it, pt, ru, tr, ja, ko, zh, id, hi, ar, vi, th

Development

npm install
npm run data    # regenerate src/data/*.json (python3 scripts/generate_data.py)
npx astro dev   # local dev server
npm run build   # static build in dist/

Built with Astro. Static site, no server: English lives at /, other locales under /<code>/.

Data pipeline: the catalog source of truth is the crawler repo (models_mapping.csv, datasets_mapping.csv), mirrored voices on Hugging Face, samples as audio files. npm run data pulls it all into src/data/*.json. See scripts/ for the tooling.

Translating

UI, guides, contact, and legal pages are translated per locale (src/i18n/). Native speakers welcome: see TRANSLATING.md to review strings and open a pull request.

Licenses

Site code: MIT (see LICENSE). Voice models, samples and datasets keep their authors' licenses; see each model page and Hugging Face repository.