AIUI - Voice Interface for AI with Multi-Model Support

AIUI: A Voice Interface for Artificial Intelligence

Introduction

The era of clicking our way through user interfaces is coming to an end. As technology advances, voice-based interfaces are poised to become the main method of interaction. AIUI is at the forefront of this transition by providing a platform for seamless, two-way verbal communication with AI models. Compatible with both desktop and mobile browsers, AIUI currently supports GPT-4 and GPT-3.5 models. The team behind AIUI is also working on compatibility with open models.

Features

How It Works

Interacting with AIUI is simple and conversational. Once you navigate to the app in your browser, you can start speaking directly to it. AIUI listens to your input, processes it using an advanced AI model, and responds with synthesized speech. This approach allows users to engage in a continuous and natural conversation, much like talking to a friend.

Running AIUI Locally

For those interested in using AIUI on their local machines, it's straightforward to set up:

Clone the Repository:

git clone [email protected]:lspahija/AIUI.git

Navigate to the AIUI Directory:
```
cd AIUI
```
Build the Docker Image:
```
docker build -t aiui .
```
If you're using an ARM64 architecture, such as Apple Silicon, use:
```
docker buildx build --platform linux/arm64 -t aiui .
```
Create a Docker Container from the Image: Run the following command, making sure to replace <YOUR_API_KEY> with your actual OpenAI API key:
```
docker run -d -e OPENAI_API_KEY=<YOUR_API_KEY> -e TTS_PROVIDER=EDGETTS -e EDGETTS_VOICE=en-US-EricNeural -p 8000:80 aiui
```
Access AIUI: Open a modern web browser and navigate to localhost:8000.

Configuration Options

The default AI model is gpt-3.5-turbo. However, you can customize it by setting the AI_COMPLETION_MODEL environment variable, choosing models like gpt-4 if your API key allows access. Language customization is also available by setting the LANGUAGE environment variable to the relevant ISO-639-1 code—English is the default. Note that languages other than English are supported when you use the gTTS or edge_tts providers for text-to-speech.

Quick Deployment

For a quick and easy deployment, AIUI offers a one-click deployment option through Railway. Users can click the deployment button provided in the project’s repository to instantly set up AIUI.

Community and Contribution

If you find AIUI valuable, contributing to its growth by starring the repository on GitHub is appreciated. This action helps contributors understand the repository’s popularity, influencing further development efforts.

AIUI is paving the way towards a voice-driven future, creating a natural and intuitive way to communicate with intelligent systems. Whether you're running it locally or deploying it with one click, AIUI offers a robust platform for those eager to embrace voice interactions with AI.