Integrate encrypted messages into audio file.
This is the course project for CUHK EE ELEG 5421 Audio signal processing
Install the environment using conda
- Installation test system: MacOS Sequoia 15.1.1
- (Since arcosx is highly restrictive, this installation should be working for most of Win & Linux systems)
conda create -n audio python=3.10
conda install matplotlib numpy pandas scikit-learn scipy
conda install -c conda-forge librosa
conda install -c conda-forge pysoundfile
conda install ipykernel
conda install pytorch torchaudio -c pytorch
conda install tqdm -c conda-forge
conda install resampy -c conda-forge
conda install huggingface_hub -c conda-forge # this might not needed nowFor wavmark, the paramaters are directly downloaded from HuggingFace: https://huggingface.co/M4869/WavMark. Parameters are stored in the folder wavmark_params.
Follow the steps in wavencrypt.ipynb