st voice

st-voice — Browse, download, and test Piper TTS voice models

Lists available TTS voices, prints download commands, and sets the active voice used by st-speak. The selected voice is saved to TTS_VOICE in your config.

Related: st-speak st-admin TTS Audio

---

Examples

st-voice --voices                   # list all available .onnx voice models
st-voice --curl                     # print curl commands to download all voices
st-voice file.txt                   # test-render a text file with the current voice
st-voice -k                         # kill the Piper TTS server on port 6419

Options

Option	Description
file.txt	Optional path to a text file to test-render with the current voice
--voices	Print names of all available Piper (.onnx) voice models
--curl	Print curl commands to download all Piper voice models
-k, --kill	Kill the Piper TTS server running on port 6419
-v, --verbose	Verbose output
-q, --quiet	Minimal output

---

For developers

Voice models (.onnx files) are discovered from the local Piper TTS server. Run st-admin to set TTS_HOST and TTS_PORT if the server is not on localhost:5000. Voice files live in ~/.cross_voices/; download them with the --curl output or set TTS_VOICE in ~/.crossenv.