Video-Transcriber

[Whisper Repo] [Colab]

Effortlessly transcribe videos and create styled subtitles with this powerful wrapper for Whisper's ASR capabilities.

Features

• Transcribe videos from various sources:
  • YouTube URLs
  • Local video files
  • Directories containing multiple videos
• Produce multiple subtitle formats:
  • SRT
  • VTT
  • TSV
  • JSON
  • Plain text
• Customize subtitle appearance with font and color options
• Leverage Whisper's advanced transcription features:
  • Language detection
  • Speech-to-text translation/transcription
  • Word level timestamps (experimental)

Requirements

• Python 3.6 or above
• yt-dlp
• moviepy
• whisper
• srt
• numpy
• torch
• ffmpeg (install with apt update && apt install ffmpeg)

Installation

1. Clone this repository:

   git clone https://github.com/abdalrohman/Video-Transcriber.git

2. Install the required dependencies:

   pip install -r requirements.txt

Usage

python video_transcriber.py [OPTIONS]

Options:

• --video_file [VIDEO_FILE]: video file input can be either a video file or directory contains videos or youtube url
• -v [VIDEO_OUTPUT_DIR]: Directory to save downloaded YouTube videos. (default: Video).
• -a [AUDIO_OUTPUT_DIR]: Directory to save extracted audio files. (default: Audio).
• -s [SRT_OUTPUT_DIR]: Directory to save generated SRT files. (default: Srt).
• --font_type [FONT_TYPE]: Font for subtitles (default: "Arial").
• --font_color [FONT_COLOR]: Color for subtitles (default: "red").
• ... (See python video_transcriber.py -h for more options)

License

This project is licensed under the MIT License. See the LICENSE file for more details.