sherpa-onnx-tts
SkillMediaA skill that gives an AI agent local text-to-speech through sherpa-onnx. It installs the sherpa-onnx runtime and a voice model on your own computer, then wraps them in a command-line tool. Your agent runs that tool with a text string and an output path to produce a WAV audio file entirely offline, with no internet or cloud service needed.
Then ask your AI: use the sherpa-onnx-tts skill
Details
Instructions available. Your AI can read the instructions. Execution depends on the setup they require.
Account requirements not reviewed. Check the skill instructions before use; Ahel provides instructions and does not run this skill.
No other account needed.
Download the sherpa-onnx runtime for your operating system and extract it into the runtime directory.
What your AI can do with it
- Turn a text string into a spoken WAV audio file
- Run text-to-speech entirely offline with no cloud service
- Use the sherpa-onnx offline CLI through a wrapped command
- Swap in other voice models from the sherpa-onnx releases
- Install the runtime and a voice model locally
Getting started
- Download the sherpa-onnx runtime for your operating system and extract it into the runtime directory.
- Download a voice model and extract it into the models directory.
- Resolve the active state directory and write the resolved runtime and model paths into the active config file.
- Set the SHERPA_ONNX_RUNTIME_DIR and SHERPA_ONNX_MODEL_DIR environment variables.
- Run the command-line tool with a text string and an output path to generate a WAV file.
What this skill tells your AI
The instructions your AI receives, as published by spinabot/brigade in skills/sherpa-onnx-tts/SKILL.md and read by Ahel’s review.
Local TTS using the sherpa-onnx offline CLI.
Install
- Download the runtime for your OS (extracts into
$BRIGADE_STATE_DIR/tools/sherpa-onnx-tts/runtime, default~/.brigade/tools/sherpa-onnx-tts/runtime) - Download a voice model (extracts into
$BRIGADE_STATE_DIR/tools/sherpa-onnx-tts/models, default~/.brigade/tools/sherpa-onnx-tts/models)
Resolve the active state directory first:
STATE_DIR="${BRIGADE_STATE_DIR:-$HOME/.brigade}"
Then write those resolved paths into the active Brigade config file ($BRIGADE_CONFIG_PATH, default ~/.brigade/brigade.json):
{
skills: {
entries: {
"sherpa-onnx-tts": {
env: {
SHERPA_ONNX_RUNTIME_DIR: "/path/to/your/state-dir/tools/sherpa-onnx-tts/runtime",
SHERPA_ONNX_MODEL_DIR: "/path/to/your/state-dir/tools/sherpa-onnx-tts/models/vits-piper-en_US-lessac-high",
},
},
},
},
}
The wrapper lives in this skill folder. Run it directly, or add the wrapper to PATH:
export PATH="{baseDir}/bin:$PATH"
Usage
{baseDir}/bin/sherpa-onnx-tts -o ./tts.wav "Hello from local TTS."
Notes:
- Pick a different model from the sherpa-onnx
tts-modelsrelease if you want another voice. - If the model dir has multiple
.onnxfiles, setSHERPA_ONNX_MODEL_FILEor pass--model-file. - You can also pass
--tokens-fileor--data-dirto override the defaults. - Windows: run
node {baseDir}\\bin\\sherpa-onnx-tts -o tts.wav "Hello from local TTS."
Signals
- GitHub stars
- 11k
- Forks
- 70
- Last commit
- Oct 2026
Questions
- Does it need an internet connection or a cloud service?
- No. The runtime and voice model are installed locally, and the command-line tool generates WAV audio files entirely offline.
- What audio format does it produce?
- It generates a WAV audio file at the output path you give it.
- Can I use a different voice?
- Yes. Other voice models can be swapped in from the sherpa-onnx releases.
- What do I need before using it?
- You need the sherpa-onnx runtime and a voice model downloaded locally, plus the SHERPA_ONNX_RUNTIME_DIR and SHERPA_ONNX_MODEL_DIR environment variables set.
Advanced
- Item type
- skill
- Key
sherpa-onnx-tts-spinabot- Source
- github.com/spinabot/brigade