Piper

Fast, lightweight local text-to-speech that even runs on a Raspberry Pi

★ Stars5.9k5,919 on GitHub
Last commit17 Oct 2026
LicenceGPL-3.0SPDX, from the repository
View on GitHub ↗

From GitHub, fetched 19 Oct 2026.

Piper is a neural text-to-speech engine with voices in dozens of languages that runs quickly on modest hardware, including a Raspberry Pi. It is maintained by the Open Home Foundation and is the default local voice for Home Assistant.

Development moved from rhasspy/piper (archived) to OHF-Voice/piper1-gpl, which is GPL-3.0.

Self-hosted alternative to

In collections

Related tools

GPT-SoVITS

Generative Artificial Intelligence (GenAI)

Voice cloning and text-to-speech from a few seconds to a minute of sample audio

Stars
★ 63.2k
Licence
MIT
Updated
  • Docker

Alternative to ElevenLabs

Kokoro-FastAPI

Generative Artificial Intelligence (GenAI)

Docker server for the natural-sounding Kokoro text-to-speech model, with an OpenAI-style API

Stars
★ 5.6k
Licence
Apache-2.0
Updated
  • Docker

Alternative to ElevenLabs

Speaches

Generative Artificial Intelligence (GenAI)

OpenAI-compatible speech server for transcription and text-to-speech

Stars
★ 3.7k
Licence
MIT
Updated
  • Docker

Alternative to OpenAI API, ElevenLabs

Ollama

Generative Artificial Intelligence (GenAI)

Get up and running with Llama 3.3, DeepSeek-R1, Phi-4, Gemma 3, and other large language models.

Stars
★ 184k
Licence
MIT
Updated
  • Docker

Alternative to ChatGPT

Open-WebUI

Generative Artificial Intelligence (GenAI)

User-friendly AI Interface, supports Ollama, OpenAI API.

Stars
★ 156k
Licence
BSD-3-Clause
Updated
  • Docker

Alternative to ChatGPT

Langflow

Generative Artificial Intelligence (GenAI)

Visual builder for AI agents and RAG workflows that you can deploy as APIs

Stars
★ 156k
Licence
MIT
Updated
  • Docker