sdatkinson/neural-amp-modeler

Features

• Audio Gear Model Trainers - Trains neural networks from paired dry and reamped audio recordings to learn gear transformations.
• Analog Gear Emulators - Trains neural networks to emulate the sound of guitar amplifiers and effects pedals.
• Audio Transfer Function Learners - Provides a supervised training pipeline using paired dry and reamped audio recordings.
• Audio Device Character Capturers - Trains neural networks on paired input-output audio samples to capture audio device sonic character.
• Audio Gear Response Modelers - Trains neural networks on paired DI and reamp audio to model guitar amplifier and effect responses.
• Neural Network Training - Trains neural networks on audio samples to replicate the tonal character of guitar amplifiers and effects.
• Neural Amp Model Runtimes - Loads serialized models and processes live audio streams with low-latency neural network inference.
• Guitar Amp Modelers - Trains neural networks to capture the tonal character of guitar amplifiers and effects pedals.
• Neural Amp Model Files - Ships a dedicated .nam file format for packaging trained neural amp models.
• Model Serialization Formats - Serializes trained neural network weights and architecture into a portable file format.
• Model Export and Portability - Exports trained neural networks into portable files that other applications can load and use.
• Model Exporting - Exports trained neural networks into file formats loadable by real-time playback plugins or standalone apps.
• Functional Model Exports - Exports trained neural networks into a file format designed for real-time playback in other applications.
• Model Export Formats - Saves trained neural networks into a portable file format that other applications can load and use.
• Model Packaging - Packages each trained neural network into a single self-contained file for simplified model management.

Star history