Jumpstart Guide To Deep Learning In Audio For Absolute Beginners: From No Experience And No Datasets To A Deployed Model
A companion repository for the Audio Developers Conference 2022 talk containing the live demo.
- The dataset for the input signal was assembled using the Fraunhofer IDMT datasets, guitar and bass in particular, thus, I cannot include it in the repository. Feel free to download the dataset from the Fraunhofer website using the form provided therein.
- The diode clipper target data was synthesized using LTspice and a model provided by Alec Wright (DiodeClipper.asc file).
- If you want to use some ready-made dataset, you can check out Alec's Guitar Amplifier Modeling repo or his Time-Varying Effects repo.