SpeechStack
Submit a template
← All templatestemplates / vonage-voice-api-websocket-connector-for-multiple-ai-engines
Customupdated Jan 28, 2025 · other · other

Vonage Voice API WebSocket Connector for Multiple AI Engines

A middleware application that connects Vonage Voice API calls (PSTN, SIP, WebRTC) to various AI engines via WebSockets. Supports inbound and outbound calling scenarios and integrates with Deepgram Voice Agent, OpenAI Realtime API, ElevenLabs Agent, Hume EVI, and custom AI pipelines through connector servers.

No demo yetView sourceFork template
The numbers
latency
cost / min
frameworkCustom
The stack
telephonyVonage
speech-to-textDeepgram Nova-2
llmGPT-4o
text-to-speechElevenLabs Turbo v2.5
System prompt
No prompt published.
Config
config.json
{
  "supports_sip": true,
  "supports_pstn": true,
  "supports_webrtc": true,
  "supports_inbound": true,
  "supports_outbound": true,
  "supported_ai_engines": [
    "Deepgram Voice Agent",
    "Deepgram ASR + OpenAI LLM + ElevenLabs TTS",
    "ElevenLabs Agent",
    "OpenAI Realtime API",
    "Hume EVI"
  ],
  "connector_architecture": true
}
Tags
websocketmiddlewareconnectorvonagetelephonypstnsipwebrtcmulti-engine
Voice Notes

Voice AI recipes, picks, and analysis.

Get the useful new templates plus the occasional teardown of what’s working in production voice AI.

contributed by @tchancse · Proprietary · source: github discoverylanguages: en-US