Piper TTS: A fast, CPU-only neural TTS engine that runs offline

The Rhasspy project's Piper TTS engine runs neural text-to-speech entirely offline on CPU-only hardware, including a Raspberry Pi 4, using a compact ~15M-parameter ONNX model that pairs a VITS-style synthesizer with the Larynx pipeline. Piper ships over 100 voices across 35+ languages at real-time speed, with the engine released under the MIT license while individual voice licenses vary, and it does not support voice cloning. The footprint and zero cloud round-trip make Piper a practical default for home automation, accessibility readers and offline kiosks where latency, privacy and hosting cost matter more than studio polish.

Piper TTS A fast, CPU-only neural TTS engine from Rhasspy that runs entirely offline, even on a Raspberry Pi. Sign up https://tts.ai/signup/ for 5,000 character limit Wrap your text in SSML tags for precise control: <speak <prosody rate="slow" Slow speech</prosody </speak Define custom pronunciations word = pronunciation : Generation Failed Commercial license from $5/mo https://tts.ai/pricing/?ref=license-note Get 200K characters every month — $5/mo https://tts.ai/checkout/?plan=lite&ref=result or a one-time 100K pack for $5 https://tts.ai/checkout/?plan=credits small&ref=result Clone a voice in 30 seconds https://tts.ai/voice-cloning/ About Piper Piper was built by the Rhasspy project to bring neural text-to-speech to edge devices that have no GPU. It pairs a VITS-style synthesizer with the Larynx pipeline and runs as a compact ~15M-parameter ONNX model directly on CPU, so synthesis happens locally with no cloud round-trip. Its strength is breadth and footprint: over 100 voices spanning 35+ languages at real-time speed on hardware as modest as a Pi 4. That makes it the practical default for home automation, accessibility readers, and offline kiosks where latency, privacy, and zero hosting cost matter more than studio polish. Best for: Quick previews, accessibility, and embedded applications Browse all Piper voices https://tts.ai/voices/?model=piper At a glance - Developer - Rhasspy - License - MIT engine ; voice licenses vary - Tier - free - Speed - fast - Voice cloning - No - Languages - English, German, French, Spanish, Italian, Portuguese, Dutch, Polish, Russian, Chinese, Arabic, Czech, Danish, Finnish, Greek, Hungarian, Icelandic, Georgian, Kazakh, Nepali, Norwegian, Romanian, Slovak, Serbian, Swedish, Swahili, Turkish, Ukrainian, Vietnamese, CA, CY, FA, Latvian, Slovenian, LB, Basque, Indonesian, Kurdish, Malayalam, Albanian, Telugu, Urdu - Max characters - 2000 Piper voices Aegis Urdu Urdu Aivars Latvian Latvian Alan UK English Alba UK English Amir Persian fa Amy US English Anna Hungarian Hungarian Antton Basque Basque Artur Slovenian Slovenian Berfin Kurdish Kurdish DFKI Turkish Turkish Darkman Polish Polish Denis Russian Russian Dutch MLS Dutch Edon Albanian Albanian Faber Portuguese Portuguese Fasih Urdu Urdu Google Nepali Nepali Gor Armenian hy Gwryw Welsh cy Harri Finnish Finnish Huayan Chinese Chinese Issai Kazakh Kazakh Jenny UK English Jirka Czech Czech Joe US English Kareem Arabic Arabic Lada Ukrainian Ukrainian Lanfrica Swahili Swahili Lessac US English Lili Slovak Slovak Maider Basque Basque Marylux Luxembourgish lb Mihai Romanian Romanian NST Swedish Swedish Natia Georgian Georgian News Indonesian Indonesian Ona Catalan ca Rapunzelina Greek Greek Riccardo Italian Italian Ryan US English Salka Icelandic Icelandic Serbian Institute Serbian Sharvard Spanish Spanish Siwis French French Tale Danish Danish Tale Norwegian Norwegian Thorsten German German VIVOS Vietnamese Vietnamese Piper TTS — FAQ ← All voices https://tts.ai/voices/