Install
openclaw plugins install openclaw-gladia@latestGladia batch speech-to-text provider for OpenClaw media understanding
openclaw plugins install openclaw-gladia@latestGladia batch speech-to-text provider for OpenClaw media understanding.
This plugin registers the gladia media-understanding provider for audio
transcription. It uses Gladia v2 pre-recorded transcription:
POST /v2/uploadPOST /v2/pre-recordedGET /v2/pre-recorded/{id} until donecd openclaw-gladia
openclaw plugins install .
openclaw gateway restart
Expose the Gladia API key to the Gateway:
GLADIA_API_KEY=gladia_xxx
Then configure audio media understanding in ~/.openclaw/openclaw.json:
{
tools: {
media: {
audio: {
enabled: true,
maxBytes: 20971520,
models: [
{
provider: "gladia",
model: "solaria-1",
language: "fr"
}
]
}
}
}
}
openclaw infer audio transcribe --file ./memo.m4a --model gladia/solaria-1 --json
Provider options can be passed through OpenClaw tools.media.audio provider
query/options. Supported primitive query fields:
languages: comma-separated language hints, for example fr,encode_switching: booleandiarization: booleansentences: booleanpunctuation_enhanced: booleanpoll_interval_ms: polling interval, default 3000npm install
npm test
npm run pack:check
This package is intentionally small: the runtime entry registers one OpenClaw
media-understanding provider named gladia, and the provider implementation
contains the Gladia v2 upload/job/polling flow.
Before publishing:
package.json once the GitHub URL exists;gladia.