py-srt-generator

A Python based Machine Learning Model that can generate SRT (subtitiles file) of any video efficiently.

How to Run

First Clone the Repo

git clone https://github.com/Mahnoor-ismail01/py-srt-generator.git
cd py-srt-generator

Next Install the required dependencies

python -m pip install -r py-srt-generator/requirements.txt

Finally, Run the script under the project root folder

python run.py

Get your subtitle file under results folder under the project root folder.
make the folder model
Download model from https://alphacephei.com/vosk/models
place it under model folder.
Then using Default model model_best.pth for generating subtitles.
Download model from https://drive.google.com/file/d/1SPwQ3tDTyUQEfAwLNQungZwXhYxA5E_I/view
Extract this model and place it under models folder.

-Change the default model path you want to use in config.py INFERENCE_PARAMS_PATH under the project root folder.

Enjoy :)

Name		Name	Last commit message	Last commit date
Latest commit History 7 Commits
__pycache__		__pycache__
csrc		csrc
data		data
jns		jns
output		output
src		src
stt		stt
temp		temp
vosk_api		vosk_api
1.srt		1.srt
Hi.srt		Hi.srt
LICENSE		LICENSE
README.md		README.md
TODO.md		TODO.md
__init__.py		__init__.py
build-dataset.ipynb		build-dataset.ipynb
config.py		config.py
decoder.py		decoder.py
encoder.py		encoder.py
erd.srt		erd.srt
fear.srt		fear.srt
fear2.srt		fear2.srt
fear3.srt		fear3.srt
fear4.srt		fear4.srt
inference.ipynb		inference.ipynb
inferer.py		inferer.py
logs.log		logs.log
logsmy.txt		logsmy.txt
post.py		post.py
postprocess.ipynb		postprocess.ipynb
qwert5.srt		qwert5.srt
r.srt		r.srt
readme.md		readme.md
requirements.txt		requirements.txt
requirements_full.txt		requirements_full.txt
run.py		run.py
t.srt		t.srt
t.srt.srt		t.srt.srt
test.srt		test.srt
train-cnn14decisionlevel.ipynb		train-cnn14decisionlevel.ipynb
utils.py		utils.py
vosktest.py		vosktest.py