Decipher provides an automated solution for generating and embedding AI-powered transcription subtitles into videos, making content more accessible. It targets video creators and editors seeking to streamline the subtitling process. The core benefit is eliminating manual transcription effort by leveraging OpenAI's Whisper model.

How It Works

Decipher utilizes OpenAI's Whisper, a state-of-the-art speech recognition system trained on a vast dataset of multilingual audio. This robust training enables Whisper to handle accents, background noise, and technical language effectively. Decipher integrates Whisper to transcribe video audio into SRT subtitle files and offers functionality to either burn these subtitles directly onto the video or use existing SRT files for subtitle embedding.

Quick Start & Requirements

• Installation: pip install git+https://github.com/dsymbol/decipher or clone and pip install . (do not use pip install decipher).
• Dependencies: Python, ffmpeg.
• Usage: decipher gui for GUI, decipher transcribe -i video.mp4 --model small for command-line transcription.
• Resources: Google Colab offers free GPU access for up to 12 hours per session.

Highlighted Details

• Leverages OpenAI's Whisper for state-of-the-art speech recognition.
• Supports automatic subtitle generation and direct burning onto videos.
• Offers a GUI via Gradio for user-friendly interaction.
• Command-line interface for transcription and subtitle embedding.

Maintenance & Community

No specific community links (Discord, Slack) or details on contributors/sponsorships are provided in the README.

Licensing & Compatibility

The README does not explicitly state the license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The project is presented as a tool for adding subtitles, but the README does not detail performance benchmarks, specific Whisper model sizes supported beyond "small," or potential limitations regarding video formats or audio quality.