CLI · open source · Pythonlive

local-tts

Make your coding agent talk to you. Offline text-to-speech from the terminal.

A tiny command-line text-to-speech tool that lets a coding agent speak its replies. Text goes in, audio comes out of a local model, and the whole thing lives in one config file rather than in one chat, so every agent on the machine talks the same way.

It began as a way to make Claude Code sound like Jarvis, in English or Spanish, without sending anything to a cloud service.

Highlights

  • Offline by defaultKokoro-82M speaks eight languages from one small model. No API key, nothing leaves the machine.
  • Zero runtime dependenciesThe package installs only itself. Standard library plus binaries you already have.
  • Tone tagsInline <calm>, <urgent>, <whisper> and more change the delivery per sentence.
  • Starts in a secondFragments play as soon as the first is synthesised instead of after the whole passage.
  • Any voice via RVCBring a trained .pth and every agent on the machine speaks in that voice, per language.
  • Six backendskokoro, piper, llama.cpp, any OpenAI-compatible endpoint, RVC, or any command that writes a WAV.
  • Built for coding agentsShips skills and a status-bar hook for Claude Code and friends, so the agent installs and drives it.
  • Language memoryRemembers which backend and voice to use for each language.

Install

git clone https://github.com/rperez93/local-tts
cd local-tts && pip install -e .
tts "<calm>Standing by.</calm>"