lightweight

Piper

GPL-3.0 engine; voice licenses vary

Tiny, fast, runs offline on a Raspberry Pi

Piper is the workhorse of local text-to-speech — built by the Rhasspy voice-assistant project to run in real time on devices as small as a Raspberry Pi. It offers a huge catalog of community voices across dozens of languages. Through the API you get the same dependable output without managing voice files or ONNX runtimes. The original MIT-licensed rhasspy/piper repo is archived; the actively maintained fork is OHF-Voice/piper1-gpl, licensed GPL-3.0.

Hear Piper

Piper 1.8.0 · en_US-ljspeech-medium · local CPU · Same three scripts across the directory.

Run it yourself

Best for: Tiny, fast, runs offline on a Raspberry Pi

Commercial conditions: Commercial use of GPL code is possible, with distribution obligations. The umbrella voice-repository tag does not license every voice. Select and review the exact voice MODEL_CARD and dataset terms before deployment.

Code: GPL-3.0 (current OHF engine). Weights: Varies by voice checkpoint. License sources checked 2026-09-20. Source 1 Source 2 Source 3

Before you choose: The current Piper engine uses GPL-3.0; each voice has its own terms. These recordings use the public-domain LJ Speech dataset voice. Older Rhasspy engine releases used MIT.

These samples were generated on an Apple Silicon CPU. No cross-model speed benchmark is claimed.

Details reviewed 2026-09-14. Hardware figures are estimates unless a benchmark is linked.

Compare with similar

Compare all →

Kokoro-82MListen side-by-side →MeloTTSListen side-by-side →KittenTTSListen side-by-side →

Choose an alternative to hear both models together.

Voices

1 voice

Female

LJ Speechdefault

American

The scripts

Neutral

“The quick brown fox jumps over the lazy dog near the riverbank.”

Emotional

“I can't believe you actually did it. This is incredible!”

Numbers & Dates

“On March 14th, 2025, the team raised $4.2 million at a 38% margin.”

Run Piper

pip install piper-tts

First audio example

Based on the author’s API. Install the documented dependencies and choose a compatible Python environment first.

python -m piper.download_voices en_US-ljspeech-medium
python -m piper -m en_US-ljspeech-medium.onnx -f output.wav -- "Hello from OpenSpeech."

Sources & setup details

Piper FAQ

Do I need a GPU to run Piper?

Piper supports CPU inference. Speed depends on your processor, runtime, and text.

Can Piper clone voices?

No — Piper uses preset voices. Browse the voice-cloning category for models that clone from reference audio.

What license is Piper released under?

Piper is released under the GPL-3.0 engine; voice licenses vary license. Always check the repository for the exact terms — some models license code and weights separately.

What languages does Piper support?

Piper supports English, Spanish, German, French, Italian, Dutch, Polish, Portuguese, Russian, UK, Chinese, Arabic, Czech, DA, EL, FI, Hungarian, NO, SV, Turkish, Vietnamese, CA, RO, SK, SR, SL.

Is there a hosted Piper API?

OpenSpeech Cloud is accepting production-access interest for Piper, but this model is not currently hosted in Cloud. Use the official repository for current deployment options.