Keras implementation of the method described in the paper 'LipNet: End-to-End Sentence-level Lipreading' by Yannis M. Assael, Brendan Shillingford, Shimon Whiteson, and Nando de Freitas (https://arxiv.org/abs/1611.01599).
- Create new conda env:
conda create -n "lipnet" python=3.7 ipython - Activate conda env:
conda activate lipnet - Install requirements:
pip install -r requirements.txt - Install ipykernel package:
conda install ipykernel --update-deps --force-reinstall - Install ffmpeg:
conda install -c conda-forge ffmpeg
- Select the Python kernel as the lipnet conda env created
- Execute all cells of the Jupyter notebook. This will download the data and model checkpoints to local directory.
- Change directory to streamlit dir:
cd app - Run streamlit app:
streamlit run streamlitapp.py --server.port 8000 - The UI can then be accessed on the browser at this link