This project provides a simple Gradio web application for transcribing YouTube videos. It targets users who need to quickly convert spoken content from YouTube videos into text, leveraging OpenAI's Whisper model for accurate speech-to-text conversion.

How It Works

The application extracts the audio stream from a given YouTube URL using the pytube library. This audio is then processed by OpenAI's Whisper model, a powerful, open-source automatic speech recognition system known for its robustness across various audio conditions and languages. The extracted text transcript is then presented to the user via the Gradio interface.

Quick Start & Requirements

• Install/Run: Clone the repository, install FFmpeg, create and activate a Conda environment (conda env create -f environment.yml, conda activate yt-whisper), then run python app.py.
• Prerequisites: Conda, Python 3.9+, FFmpeg (must be in system PATH or specified manually).
• Setup: Requires cloning the repo, installing FFmpeg, and setting up a Conda environment.
• Links: GitHub Repository

Highlighted Details

• Utilizes OpenAI's Whisper model for transcription.
• Built as a Gradio application for an interactive web interface.
• Handles audio extraction from YouTube videos.

Maintenance & Community

No specific information on maintainers, community channels, or roadmap is provided in the README.

Licensing & Compatibility

• License: MIT License.
• Compatibility: Permissive license suitable for commercial use and integration with closed-source projects.

Limitations & Caveats

The application relies on pytube for YouTube access, which can be subject to breakage due to YouTube's frequent changes. FFmpeg must be correctly installed and accessible in the system's PATH for the application to function.