---
title: "Swift – AI Voice Assistant"
description: "Fast voice assistant powered by Groq, Cartesia, VAD, and Vercel."
url: "https://vercel.com/templates/next.js/swift-ai-voice-assistant"
links:
  repository: "https://github.com/ai-ng/swift"
  demo: "https://swift-ai.vercel.app"
  deploy: "https://vercel.com/new/clone?demo-title=Swift+%E2%80%93+AI+Voice+Assistant&demo-description=Fast+voice+assistant+powered+by+Groq%2C+Cartesia%2C+VAD%2C+and+Vercel.&demo-url=https%3A%2F%2Fswift-ai.vercel.app&demo-image=%2F%2Fimages.ctfassets.net%2Fe5382hct74si%2F1ioGQ6QcT5TgZQ9yWJMwfE%2Ff9231142678dd0ce1d8f217b24d0afb2%2F350604530-362c939a-790b-4c1a-a17b-b8e7858507ce.png&project-name=Swift+%E2%80%93+AI+Voice+Assistant&repository-name=swift-ai-voice-assistant&repository-url=https%3A%2F%2Fgithub.com%2Fai-ng%2Fswift&from=templates&skippable-integrations=1&env=GROQ_API_KEY%2CCARTESIA_API_KEY"
---

# Swift – AI Voice Assistant

Fast voice assistant powered by Groq, Cartesia, VAD, and Vercel.

# [Swift](https://swift-ai.vercel.app)

Swift is a fast AI voice assistant.

-   [Groq](https://groq.com) is used for fast inference of [OpenAI Whisper](https://github.com/openai/whisper) (for transcription) and [Meta Llama 3](https://llama.meta.com/llama3/) (for generating the text response).
-   [Cartesia](https://cartesia.ai)'s [Sonic](https://cartesia.ai/sonic) voice model is used for fast speech synthesis, which is streamed to the frontend.
-   [VAD](https://www.vad.ricky0123.com/) is used to detect when the user is talking, and run callbacks on speech segments.
-   The app is a [Next.js](https://nextjs.org) project written in TypeScript and deployed to [Vercel](https://vercel.com).

Thank you to the teams at Groq and Cartesia for providing access to their APIs for this demo!

[![Deploy with Vercel](https://vercel.com/button)](https://vercel.com/new/clone?repository-url=https%3A%2F%2Fgithub.com%2Fai-ng%2Fswift&env=GROQ_API_KEY,CARTESIA_API_KEY&envDescription=Groq%20and%20Cartesia's%20APIs%20are%20used%20for%20transcription%2C%20text%20generation%2C%20and%20speech%20synthesis.&project-name=swift&repository-name=swift&demo-title=Swift&demo-description=A%20fast%2C%20open-source%20voice%20assistant%20powered%20by%20Groq%2C%20Cartesia%2C%20and%20Vercel.&demo-url=https%3A%2F%2Fswift-ai.vercel.app&demo-image=https%3A%2F%2Fswift-ai.vercel.app%2Fopengraph-image.png)

## Developing

-   Clone the repository
-   Copy `.env.example` to `.env.local` and fill in the environment variables.
-   Run `pnpm install` to install dependencies.
-   Run `pnpm dev` to start the development server.

## Related Templates

### [Hume AI - Empathic Voice Interface Starter](https://vercel.com/templates/next.js/empathic-voice-interface-starter)

This template creates a voice chat using Hume AI's Empathic Voice Interface.

### [RAGBot Starter — An Astra DB and OpenAI chatbot](https://vercel.com/templates/next.js/ragbot-starter)

Starter project for creating a chatbot using Astra DB and OpenAI.

### [Chatbot](https://vercel.com/templates/next.js/chatbot)

A full-featured, hackable Next.js AI chatbot built by Vercel

---

## Additional documentation

The [Vercel Knowledge Base](https://vercel.com/kb) contains guides and answers to common questions about Vercel, including deployment, framework configuration, domains, caching, and troubleshooting.