Skip to content

A Streamlit based application that helps you chat with your audio file, powered by Langchain, ChromaDB, and OpenAI. Used to chat with any YouTube Video by inputting the URL into a pipeline that extracts the video transcript and feeds it into the GPT-4 .

Notifications You must be signed in to change notification settings

calicartels/LLMTalk

Repository files navigation

Chat-with-Audio-using-LLM

This is a Streamlit application that helps you chat with your audio file powered by Langchain, ChromaDB, and OpenAI.

An app to help people skip the nuances of watching the whole video

IntroductionInstallationApplication StructureKey FeaturesHow To UseReferences

Introduction

The repository contains the code for a Streamlit application that allows users to chat with YouTube video files. The application is powered by large language models (LLMs) and Vector databases. It can be used to learn more about audio files, create new and creative ways to interact with Audio/Video content and explore the intersection of AI and Multimedia.

Installation

Install with pip:

$ pip install -r requirements.txt

Application Structure

App

Key Features

⭐️ Supports a variety of audio/video formats, including WAV, MP3, and FLAC.
⭐️ Can generate transcripts on its own.
⭐️ Can be used to create new and creative ways to interact with audio content.

How To Use

To run this application you will need an OpenAI API Key

From your command line:

# Clone this repository
$ git clone https://github.com/calicartels/LLMTalk

# Go into the repository
$ cd LLMTalk


# Install dependencies
$ pip install -r requirements.txt

References

  1. https://www.youtube.com/@AIAnytime
  2. https://towardsdatascience.com/getting-started-with-streamlit-web-based-applications-626095135cb8
  3. https://blog.logrocket.com/implement-vector-database-ai/

About

A Streamlit based application that helps you chat with your audio file, powered by Langchain, ChromaDB, and OpenAI. Used to chat with any YouTube Video by inputting the URL into a pipeline that extracts the video transcript and feeds it into the GPT-4 .

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages