openai/openai-voice-agent-sdk-sample

openai/openai-voice-agent-sdk-sample

Description: Sample application to add voice capabilities to the Agents SDK

Language: TypeScript

License: MIT

Stars: 257

Forks: 74

Open issues: 1

Created: 2025-04-08T14:33:08Z

Pushed: 2025-05-13T05:25:08Z

Default branch: main

Fork: no

Archived: no

README:

Voice Agents SDK Sample App

This repository contains a sample app to highlight how to build voice agents using the Agents SDK and Python. The backend is written using FastAPI and exposes a websocket endpoint. The front-end is written using Next.js and connects to the websocket server.

Features:

• Multi-turn conversation handling
• Push-to-talk audio mode
• Function calling
• Streaming responses & tool calls

This app is meant to be used as a starting point to build a conversational assistant that you can customize to your needs.

Requirements

• OpenAI API key
• If you're new to the OpenAI API, sign up for an account.
• Follow the Quickstart to retrieve your API key.
• Node.js and npm
• uv installed on your system

How to use

1. Set the OpenAI API key:

2 options:

• Set the OPENAI_API_KEY environment variable globally in your system
• Set the OPENAI_API_KEY environment variable in the project: Create a .env file at the root of the project and add the following line (see .env.example for reference):

OPENAI_API_KEY=

2. Clone the Repository:

git clone https://github.com/openai/openai-voice-agent-sdk-sample.git
cd openai-voice-agent-sdk-sample/

3. Install dependencies:

You will have to install both the dependencies for the front-end and the server. To do this run in the project root:

make sync

4. Run the app:

make serve

The app will be available at `http://localhost:3000`.

Contributing

You are welcome to open issues or submit PRs to improve this app, however, please note that we may not review all suggestions.

License

This project is licensed under the MIT License. See the LICENSE file for details.