An implementation of speech reconstruction methods from Deep Image Prior (Ulyanov et al., 2017) in PyTorch.
The point of the project is to verify speech enhancement task using neural networks untrained on data prior to use.
Python3 with Python3 Tensorflow, and NumPy. CUDA and cuDNN are optional (settable within the script in a self-explanatory way) but strongly recommended.
TODO soon
mkdir output
Generate output speech with:
python3 deep_speech_prior.py