devilsnerve/ai-transcribe-mp4

Automate audio-to-text transcription with Python. This tool downloads videos, extracts audio, transcribes using OpenAI's Whisper, and outputs PDFs. Features multithreading for efficiency.

---

Audio to Text Transcription Tool

Overview

This tool is designed to automate the process of converting audio from a video into a text transcription. It downloads a video, extracts the audio, splits the audio into manageable segments, transcribes each segment using OpenAI's Whisper model, and compiles the transcriptions into a single PDF file.

Requirements

• Python 3.x
• requests
• pydub
• moviepy
• FPDF
• tqdm
• whisper
• concurrent.futures
• multiprocessing
• reportlab

Installation

Before running the script, ensure you have Python installed on your machine and then install the required packages using pip:

pip install requests pydub moviepy fpdf tqdm whisper reportlab

Usage

To use the tool, follow these steps:

1. Ensure ffmpeg.exe is accessible in your system's PATH.
2. Modify the video_url variable in the script to point to the video you want to transcribe.
3. Run the script using Python.

Example

Here's an example of how to run the script:

python transcription_tool.py

The script performs the following steps:

1. Downloads the video from the specified URL.
2. Extracts the audio from the downloaded video.
3. Splits the audio into segments and saves them in a directory.
4. Transcribes each audio segment using Whisper and saves the transcriptions.
5. Compiles all transcriptions into a PDF file.

Notes

• The script uses OpenAI's Whisper model for transcription, which requires a good internet connection for model downloading.
• The script currently saves the transcribed segments in a folder named 'transcriptions'.

---