This is a Node.js application that demonstrates the audio capabilities of LiteLLM proxy. Users can speak into their microphone or type text, which is then sent to the LiteLLM proxy for processing. The application plays back the audio response. Note: This version is not optimal, you should use the streaming version of the liteLLM API

Features

• Voice input through microphone
• Text input option
• Speech-to-text using LiteLLM's transcription service
• Text-to-speech using LiteLLM's speech service
• Conversation history display

Prerequisites

• Node.js (v14 or higher recommended)
• LiteLLM proxy API key

Installation

1. Clone this repository

git clone <https://gitlab.ethz.ch/sph/digital-makerspace/lite-llm-audio-demo.git>
cd lite-llm-audio-demo

1. Install dependencies

npm install

1. Set up your LiteLLM API key
   1. Create a .env file in the root directory
   2. Add your API key:
      • Create a .env file in the root directory
      • Add your API key: LITELLM_API_KEY=your-api-key-here

Running the Application

1. Start the server:

node server.js