speech-engine

Give your custom agent a real-time voice.

voice agents

What it does

Adds real-time voice conversations to a custom agent using a developer-hosted server and ElevenLabs speech services.

When to use it

Use when adding voice to an agent or chat app, hosting its WebSocket server, streaming spoken responses, or handling interruptions.

Give it a custom agent project; it implements the Speech Engine server, browser token endpoint, and voice client integration.

What you provide

This skill

ElevenLabs

Write

Sends voice data to ElevenLabs

Speech Engine server

Execute

Starts the Speech Engine server

microphone

Read

Captures audio from your microphone

Reads the ElevenLabs API key from the server environment for SDK authentication.