The project

PyTorch is a well-designed, flexible, popular, and well-documented toolkit with a very large community. Most speech applications rely on deep learning and signal processing techniques, that can be naturally implemented in PyTorch. Processing steps are performed either on GPUs or CPUs. It is feasible to design end-to-end differentiable systems, where the gradient can potentially flow through all the different parts of the architecture, including parts solving different audio and speech tasks (*e.g. PyTorch is a well-designed, flexible, popular, and well-documented toolkit with a very large community. Most speech applications rely on deep learning and signal processing techniques, that can be naturally implemented in PyTorch.