Menu
Real-time streaming speech-to-text via Deepgram WebSocket API
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Edit, transform, and enhance images using AI models
Extract text from images using OCR and vision AI
Neural text-to-speech via Amazon Polly
Speaker diarization — identifies and tracks who is speaking at each moment in an audio stream
Semantic endpoint detection — uses an LLM to classify whether the user's utterance is a complete thought, reducing false turn boundaries on mid-sentence pauses
Batch speech-to-text via Google Cloud Speech-to-Text API
| name | streaming-stt-deepgram |
| description | Real-time streaming speech-to-text via Deepgram WebSocket API |
| category | voice |
Real-time streaming speech-to-text using Deepgram's Nova-2 model via WebSocket.
Set DEEPGRAM_API_KEY in your environment or agent secrets store.
speaker labels)providerOptions.keywordsIn agent.config.json:
{
"voice": {
"stt": "deepgram"
}
}
Provider-specific options via providerOptions:
{
"voice": {
"stt": "deepgram",
"providerOptions": {
"model": "nova-2",
"diarize": true,
"keywords": ["AgentOS:2", "Deepgram:1.5"],
"endpointing": 300
}
}
}
| Event | Payload | Description |
|---|---|---|
transcript | TranscriptEvent | Every hypothesis (interim + final) |
interim_transcript | TranscriptEvent | Non-final hypothesis |
final_transcript | TranscriptEvent | Stable, final hypothesis |
speech_start | — | First non-empty word in an utterance |
speech_end | — | Deepgram speech_final flag raised |
error | Error | Unrecoverable provider error |
close | — | Session fully terminated |