speech-engine

Name: speech-engine
Availability: InStock
Author: ri-ru

Integrate ElevenLabs real-time voice streaming into custom agent runtimes via WebSocket.

Updated Jul 18, 2026

Search Tags:#voice #text-to-speech #websockets #elevenlabs #speech-to-text #webrtc

Ready to run

npx skills add https://github.com/ri-ru/cyrano --skill speech-engine

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: speech-engine
Source: https://github.com/ri-ru/cyrano/tree/main/.agents/skills/speech-engine
Command: npx skills add https://github.com/ri-ru/cyrano --skill speech-engine

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @elevenlabs/elevenlabs-js, @elevenlabs/react, @elevenlabs/client, dotenv, express, tsx, elevenlabs, python-dotenv, flask, and includes references (resource) components.

What problem does it solve?

This skill solves the complexity of integrating real-time, low-latency voice interactions into custom agent runtimes, handling the heavy lifting of WebSockets, turn-taking, and audio streaming.

Core Features & Use Cases

Real-time Voice Interface: Connects your server logic to ElevenLabs for speech-to-text and text-to-speech.
Interruption Handling: Manages user interruptions during streaming responses to ensure natural conversation flow.
Use Case: Build a voice-enabled customer support agent that validates user intent on your own server before streaming a response back to the browser.

Quick Start

Use the speech-engine skill to initialize a new WebSocket server endpoint that connects your agent logic to the ElevenLabs voice platform.

Frequently Asked Questions about speech-engine

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I build a real-time voice agent with ElevenLabs speech-to-text?▼

To build a real-time voice agent with ElevenLabs, this Skill integrates WebSocket protocols for bidirectional audio streaming and turn-taking management. It connects your custom server logic to ElevenLabs for speech-to-text and text-to-speech processing.

Can I handle user interruptions during streaming audio responses?▼

Yes, you can handle user interruptions during streaming audio responses. The Skill manages turn-taking logic to detect user speech and stop ongoing text-to-speech playback, ensuring a natural conversational flow without overlapping audio.

What do I need to set up a WebSocket server for real-time voice interfaces?▼

To set up a WebSocket server for real-time voice interfaces, you need ElevenLabs API credentials and a public HTTPS WebSocket URL. These establish the secure communication channels required for bidirectional audio streaming.

Does this Skill support server-side response validation for voice applications?▼

Yes, this Skill supports server-side response validation for voice applications. You can validate user intent on your own server before streaming the text-to-speech response back to the browser through the WebSocket connection.

How does bidirectional audio streaming work with WebSockets and ElevenLabs?▼

Bidirectional audio streaming works by connecting your server to ElevenLabs Speech Engine via WebSocket protocols. This facilitates simultaneous speech-to-text transcription of user audio and text-to-speech streaming of agent responses.

Are there limitations when integrating voice capabilities into custom agent runtimes?▼

A key limitation when integrating voice capabilities into custom agent runtimes is the strict requirement for a public HTTPS WebSocket URL and valid ElevenLabs API credentials to establish secure communication channels for audio streaming.