Subtitle generation tool (Web-UI + CLI + Python package) using Whisper
Top 27.6% on sourcepulse
Subs AI is a comprehensive tool for generating subtitles from audio and video files, targeting developers and end-users alike. It leverages OpenAI's Whisper and its variants, offering transcription, translation, and subtitle manipulation capabilities through a Web UI, CLI, and Python package.
How It Works
The project integrates multiple Whisper implementations, including OpenAI's original, whisper-timestamped
for word-level timestamps, whisper.cpp
for CPU inference, faster-whisper
for optimized performance (up to 4x faster with quantization), and whisperX
for high-speed transcription with word-level timestamps and speaker diarization. It also supports stable-ts
for timestamp stabilization and Hugging Face Transformers for broader model compatibility. Translation is handled by NLLB and mBART models.
Quick Start & Requirements
pip install git+https://github.com/absadiki/subsai
ffmpeg
to be installed.Highlighted Details
whisperX
.ffsubsync
..srt
, .vtt
, .ass
, etc.).Maintenance & Community
Licensing & Compatibility
Limitations & Caveats
Python 3.12+ may have compatibility issues. The GPLv3 license imposes significant restrictions on commercial use and integration with proprietary software.
1 week ago
Inactive