A GPU-optional modular synthesizer in pytorch
torchsynth torchsynth is based upon traditional modular synthesis written in pytorch. It is GPU-optional and differentiable. Most synthesizers are fast in terms of latency. torchsynth is fast in terms of throughput. It synthesizes audio 16200x faster than realtime (714MHz) on a single GPU. This is of particular interest to audio ML researchers seeking large training corpora. Additionally, all synthesized audio is returned with the underlying latent parameters used for generating the corresponding audio. This is useful for multi-modal training regimes. […]
Read more