Skip to content

Repository files navigation

Sirene

CI Release Docs

Self-hosted multi-backend text-to-speech platform with voice cloning and a modern web UI.

Full documentation: kevinbonnoron.github.io/sirene

Screenshots

Login

Dashboard Voices
Models History

Quick Start

curl -sSL https://raw.githubusercontent.com/KevinBonnoron/sirene/main/install.sh | bash

Then open http://localhost.

Features

  • Multi-backend TTS - Route requests to Kokoro, Qwen3-TTS, F5-TTS, Piper, CosyVoice, Fish Speech, or Chatterbox from a single interface
  • Voice cloning - Create custom voices by uploading audio samples with zero-shot cloning
  • Model management - Download and manage TTS models on demand from the web UI
  • Real-time updates - Track downloads and generation progress via Server-Sent Events
  • Transcription - Speech-to-text via Whisper models
  • Self-hosted - Two Docker images: a single Go binary for the web/API/database, and one for inference

Supported Backends

Backend Voice Cloning Streaming Languages
Kokoro - - EN, FR, JA, KO, ZH
Qwen3-TTS Yes - 10+ languages
F5-TTS Yes - Multilingual
Piper - - 26 languages
CosyVoice Yes Yes 9 languages
Fish Speech S2-Pro Yes - 12+ languages
Chatterbox Yes - EN + 23 languages
Higgs Audio V2 Yes - Multilingual
Voxtral (Mistral) Yes Yes 9 languages

Speed and pitch are available on every backend above: those that expose no native control are adjusted on the rendered audio by the worker. The ElevenLabs and OpenAI voices are generated by their provider and never reach the worker, so they take speed only.

Development

Prerequisites

  • Task (task runner)
  • Bun >= 1.3.10
  • Go >= 1.26
  • Python >= 3.11
  • For the desktop app on Linux: GTK 3 and WebKitGTK 4.1 development headers

Quick Start

With Nix, direnv allow (or nix develop) provides every tool. For manual setup:

task install
pip install -e "./inference[cpu]"

Start all services

task dev
Service Port
Sirene server (API + PocketBase, dashboard at /_/) 8090
Vite Client 5173
Inference FastAPI 8000

Tasks

task --list        # Every task
task dev           # Server, client and inference worker
task build         # Web UI, server binary and CLI
task build:ui      # Build the client into server/ui/dist for embedding
task test          # Go, client and inference test suites
task lint          # Biome, gofmt, go vet, ruff
task desktop:build # Wails desktop app for the host OS
task desktop:package

License

MIT

About

Self-hosted text-to-speech platform with multi-backend support, voice cloning, and a modern web UI.

Topics

Resources

Stars

13 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages

Generated from KevinBonnoron/template