Text-to-speech
Piper (voices) requirements, license & speed
By Rhasspy / Open Home Foundation · released 2023-06-22
Commercial useConditional
Min memoryNot stated
Runs on CPUYes
StreamingUnknown
Key facts
- License
- Voices repo: MIT; engine (piper1-gpl): GPL-3.0 source
- Commercial use
- Conditional sourceEngine is GPL-3.0; each voice's training data has its own license, check the voice's MODEL_CARD
- Parameters
- Unknown
- Minimum memory
- Not stated by source
- Runs on CPU
- Yes sourceFast local neural TTS, ONNX on CPU (smoke-tested here)
- Streaming
- Unknown
- Voice cloning
- Unknown
- Languages
- ar, ca, cs, cy, da, de, el, en, es, fa, fi, fr, hu, is, it, ka, kk, lb, lv, ne, nl, no, pl, pt, ro, ru, sk, sl, sr, sv, sw, tr, uk, vi, zh source
- Released
- 2023-06-22 sourceHugging Face repo creation date (proxy for release date)
- Gated on HF
- No
Measured speed & memory
| Hardware | Runtime / settings | RTF | Time to first audio | Peak RAM | Date |
|---|---|---|---|---|---|
| Cloud x86 VM, 8 vCPU Intel Xeon, 16 GB RAM, no GPU | piper-tts (onnxruntime) voice=en_US-lessac-medium | 0.0174 (57× real time) | 0.051 s | 378.9 MB | 2026-10-21 |
Median of 5 runs after one warm-up, ~10 s English passage. Hardware is labelled exactly: a cloud x86 VM is not an N100 or a Pi. Full method on the CPU only page. Peak RAM is whole-process RSS. Methodology
Accuracy
We don't run accuracy evals. For quality and intelligibility, check the Hugging Face Open TTS Leaderboard.
Comparisons: Kokoro vs Piper
Similar TTS models
Spotted an error? Report it. Every fact links to the model card or repo it came from.