Skip to content
Get packs

Local Text-to-Speech

Turn text into spoken audio offline using sherpa-onnx voice models on your own machine

Source Download skill
Repository
spinabot/brigade
Stars
11,367
Licence
MIT
Files
1 file

What it does

This skill lets an AI agent generate spoken audio files from text without sending anything to a cloud service. Once installed, the agent can run a bundled wrapper command that uses the open-source sherpa-onnx speech engine and a downloadable voice model to write a WAV file on your computer. Setup involves downloading a runtime and a voice model and pointing the skill at them through environment variables.

Great for

  • Creators who want narration or voiceover drafts produced privately on their own machine
  • Developers building apps or agents that need offline speech output
  • People who want to avoid per-request fees or sending text to a third-party TTS provider

Skip it if

  • Users who need high-end, emotionally expressive or multilingual voices, since this skill is set up around a single English (US) Piper voice model
  • Users on unsupported systems, since the skill only lists macOS, Linux and Windows and downloads x64 builds for Linux and Windows, and a universal build for macOS
  • Anyone expecting a one-click install: it downloads large archives from GitHub and requires editing a config file and setting paths

Try asking

  • “Make a WAV file saying 'Welcome to our product tour' using local text-to-speech”
  • “Generate a short voiceover for this paragraph and save it as narration.wav”
  • “Set up offline TTS with the default voice so my agent can read out notifications”
Step-by-step guide: how to use Local Text-to-SpeechInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    sherpa-onnx-tts.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip sherpa-onnx-tts.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip sherpa-onnx-tts.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Review the download URLs and the bundled bin/sherpa-onnx-tts wrapper before running, since the agent will execute it and the install step fetches binaries from GitHub
  • Pin or verify the runtime and model versions yourself if you need a reproducible setup, and keep the paths in the config file in sync with your state directory
  • If you add more voice models to the model folder, set SHERPA_ONNX_MODEL_FILE or pass --model-file to avoid ambiguity

What's inside

  • SKILL.md

Redistributed unmodified from spinabot/brigade under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.