Skip to content

Latest commit

 

History

History
96 lines (55 loc) · 2.31 KB

README.md

File metadata and controls

96 lines (55 loc) · 2.31 KB
Raw

AI subtitle Editor

The final goal of this project is turn the subtitle into a readable written file with slids by using AI, that is Large Language Model (LLM) promts.

This is my personal project for GCDFL. The lectures will generate the cc.vtt or srt files. However, I want to make these files into readable markdown file for Chinese audience.

you can try the demo, right now only remove the timestamps and joining the lines.

Content

Features

Install

Features:

  • Remove transcript timestamps and joining the lines. [Done]

  • delete the oral words, such hi, a, ha etc.

  • delete repeated words, phrase and sentences.

  • more.

Install

prerequest

  • install rye
  • whisper.cpp (optional, if you want use the a_w_whisper-cpp.py, recommend for cpu user.)
  • whisper (optional, if you want use the a_w-whisper.py)

first step clone this repository

git clone https://github.com/Areopaguaworkshop/AI-Subtitle-Editor.git

second step

cd AI-Subtitle-Editor

mv pyproject.toml pyproject-bk.toml

rye init

third step

copy whole content of the pyproject-bk.toml into pyproject.toml

Then run

rye sync

four step

python src/rm_time_join_line.py

  • if you want local whisper with audio transcribe, you can run

python src/a_w-whisper

or

python src/a_w_cpp.py # you will need to dowload the whispercpp.bin

or, if you have the GROQ_API_KEY, you can first

export GROQ_API_KEY="put your GROQ_API_KEY here"

then run

python src/a_w_groq.py

Enjoy!

Buy me a Cofee.

License:

AI-Subtitle-Editor is licensed under the Apache License 2.0 found in the LICENSE file in the root directory of this repository.

Citation:

  title = {AI-Subtitle-Editor},
  author = {Yuan, Yongjia},
  year = {2024},
}