Vibeleaderboard
← Back to Tools
Index / tool

loudkit

github.com/loudreader/loudkit
Visit github.com
Category
AI Tools
Rank
No. 1146Tools index
Pricing
Open Source
Platform
cli · desktop · web · mobile
Type
TOOL
GitHub
10 stars
Latest release
v0.1.1
Date

About

An open-source, offline text-to-speech engine offering 28 voices across 10 languages with voice cloning from ~10 seconds of audio, plus native SDKs for Python, Swift, Go, Rust and TypeScript. It ships a CLI, an HTTP/gRPC server with an OpenAI-compatible speech API, and a preview MCP server for agent integration, all running locally with no account or usage fees.

What it can do

  • Convert text to speech offline

    Text → Audio speech in one of 28 voices/10 languages

  • Clone a voice from a short audio sample

    ~10 seconds of audio → Cloned voice model for TTS

  • Run text-to-speech via command-line interface

    Text and CLI commands → Generated audio file

  • Serve text-to-speech requests over HTTP/gRPC with an OpenAI-compatible speech API

    API requests → Synthesized speech audio

  • Integrate with AI agents via a preview MCP server

    MCP requests → Speech output for agent workflows

  • Perform text-to-speech through native SDKs in Python, Swift, Go, Rust, and TypeScript

    Text via SDK calls → Generated audio

Why it made the leaderboard

Running TTS entirely on-device with native SDKs across five languages/ecosystems removes the latency, cost, and privacy tradeoffs of cloud TTS APIs, useful for developers building voice features that need to work offline.

Tags

text-to-speechvoice-cloningofflinelocal-aionnxmultilingualttsmcp

Tech Stack

PythonSwiftDocker

Comments (0)

No comments yet

Editorially curated, with community endorsements as a secondary signal. Corrections welcome.