# Parametric Neural Amp Modeling with Active Learning

arXiv: https://arxiv.org/abs/2509.26564 · HTML: https://mlsanity.com/paper/2509.26564/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 30 references that are in the map.
- First posted on arXiv 30 Sep 2025.

## Builds on (most cited first, 12 of 12)

- [Simple and Scalable Predictive Uncertainty Estimation using Deep Ensembles](https://mlsanity.com/paper/1612.01474/) (2016)
- [WaveNet: A Generative Model for Raw Audio](https://mlsanity.com/paper/1609.03499/) (2016)
- [High-Fidelity Audio Compression with Improved RVQGAN](https://mlsanity.com/paper/2306.06546/) (2023)
- [Deep Learning for Tube Amplifier Emulation](https://mlsanity.com/paper/1811.00334/) (2018)
- [Efficient neural networks for real-time modeling of analog dynamic range compression](https://mlsanity.com/paper/2102.06200/) (2021)
- [Modelling black-box audio effects with time-varying feature modulation](https://mlsanity.com/paper/2211.00497/) (2022)
- [A general-purpose deep learning approach to model time-varying audio effects](https://mlsanity.com/paper/1905.06148/) (2019)
- [Sample Rate Independent Recurrent Neural Networks for Audio Effects Processing](https://mlsanity.com/paper/2406.06293/) (2024)
- [Differentiable Black-box and Gray-box Modeling of Nonlinear Audio Effects](https://mlsanity.com/paper/2502.14405/) (2025)
- [End-to-End Amp Modeling: From Data to Controllable Guitar Amplifier Models](https://mlsanity.com/paper/2403.08559/) (2024)
- [Real Time Emulation of Parametric Guitar Tube Amplifier With Long Short Term Memory Neural Network](https://mlsanity.com/paper/1804.07145/) (2018)
- [Hyper Recurrent Neural Network: Condition Mechanisms for Black-box Audio Effect Modeling](https://mlsanity.com/paper/2408.04829/) (2024)

## Abstract (arXiv)

We introduce Panama, an active learning framework to train parametric guitar amp models end-to-end using a combination of an LSTM model and a WaveNet-like architecture. With model, one can create a virtual amp by recording samples that are determined through an ensemble-based active learning strategy to minimize the amount of datapoints needed (i.e., amp knob settings). Our strategy uses gradient-based optimization to maximize the disagreement among ensemble models, in order to identify the most informative datapoints. MUSHRA listening tests reveal that, with 75 datapoints, our models are able to match the perceptual quality of NAM, the leading open-source non-parametric amp modeler.