speedlearn-cli

speedlearn-cli is a command-line tool that automates the process of downloading MP4 videos, converting them to MP3, transcribing the audio content, and generating detailed notes. This project was entirely built with Claude 3.5 Sonnet with Cursor in under an hour. Not a single line of code was written by a human.

Resources

Features

• Download MP4 videos and convert them to MP3 format
• Transcribe audio files using the Groq API (OpenAI Whisper Large-v3 Turbo model)
• Generate detailed notes and minutia using the Anthropic Claude API (NEW Claude 3.5 Sonnet model)
• Efficient file handling with caching to avoid redundant operations

Prerequisites

• Go 1.x or higher
• FFmpeg (for video to audio conversion)
• Groq API key
• Anthropic API key

Installation

1. Clone the repository:

   git clone https://github.com/yourusername/speedlearn-cli.git
   cd speedlearn-cli
   

2. Install dependencies:

   make deps
   

3. Build the project:

   make build
   

Configuration

Create a .env file in the project root and add your Groq API key:

Usage

Run the tool with the following command: ./speedlearn-cli --filename <output_filename> --url <video_url>

The tool will:

1. Download the video and convert it to MP3
2. Transcribe the audio using the Groq API
3. Save the MP3 and transcription in the output/<filename> directory

Project Structure

• main.go: Main application logic and CLI interface
• convert.go: Functions for video download and conversion to MP3
• transcribe.go: Functions for audio transcription using the Groq API
• notes.go: Functions for generating notes using the Anthropic Claude API
• Makefile: Build and management commands

Development

• Run tests: make test
• Clean build artifacts: make clean
• Build for Linux: make build-linux

Note

This entire project was conceptualized and implemented with the assistance of Claude 3.5 Sonnet, an AI language model, in under an hour. While functional, it may require further testing and refinement.