Skip to content

fzinnah17/Yoda

Repository files navigation

Yoda: Conversation Co-Pilot

This repository contains the code for Yoda, a Conversation Co-Pilot application that utilizes Gradio for audio transcription and response generation based on OpenAI models.

Table of Contents

About

Yoda is a Conversation Co-Pilot that assists users in transcribing spoken audio and generating context-aware, polite, and assertive responses to facilitate workplace conversations.

Features

  • Audio transcription using OpenAI's Whisper ASR model.
  • Response generation for workplace scenarios using OpenAI's GPT-based models.
  • Real-time interaction through Gradio's user-friendly interface.

Stack

  • Hugging Face (HugginFace) for transformer models
  • Whisper for Automatic Speech Recognition (ASR)
  • TogetherAI for collaborative features
  • LangChain for language processing

Prerequisites

Ensure you have the following prerequisites installed before using Yoda:

  • Python
  • Gradio
  • Transformers library
  • OpenAI API key

Installation

Clone the repository: git clone https://github.com/fzinnah17/Yoda

Install the required dependencies: pip install -r requirements.txt

Set your Open AI API Key: export OPENAI_API_KEY = "your-api-key-here"

Run Yoda, the Conversation Co-Pilot: python app.py

Demo

Yoda.-.Conversation.Copilot.mov

Slide Deck

Pitch Deck

Usage

Launch Yoda by running the provided command. Speak into the microphone to trigger audio transcription. Experience real-time, context-aware response generation based on workplace scenarios.

Future Work

The future work for Yoda, the Conversation Co-Pilot, includes:

  • Implementing a vector database to store entire conversations.
  • Allowing the Language Model (LLM) to reference back to the stored conversations.
  • Summarizing chat history and providing access to the most recent dialogue.

Contributing

If you like to contribute to Yoda, the Conversation Co-Pilot, please follow the guidelines outlined in the CONTRIBUTING.md file.

License

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

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages