Deep Learning for Tube Amplifier Emulation

@article{Damskgg2019DeepLF,
  title={Deep Learning for Tube Amplifier Emulation},
  author={Eero-Pekka Damsk{\"a}gg and Lauri Juvela and Etienne Thuillier and Vesa V{\"a}lim{\"a}ki},
  journal={ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  year={2019},
  pages={471-475}
}
Analog audio effects and synthesizers often owe their distinct sound to circuit nonlinearities. [] Key Method Specifically, a feedforward variant of the WaveNet deep neural network is trained to carry out a regression on audio waveform samples from input to output of a SPICE model of the tube amplifier. The output signals are pre-emphasized to assist the model at learning the high-frequency content. The results of a listening test suggest that the proposed model accurately emulates the reference device. In…

Figures and Tables from this paper

Real-Time Guitar Amplifier Emulation with Deep Learning
TLDR
It is demonstrated that the neural network models can convincingly emulate highly nonlinear audio distortion circuits, whilst running in real-time, with some models requiring only a relatively small amount of processing power to run on a modern desktop computer.
Real-Time Modeling of Audio Distortion Circuits with Deep Learning
TLDR
The deep neural network studied in this work is useful for real-time virtual analog modeling of nonlinear audio circuits and can be run in real time on a modern desktop computer.
Deep Learning for Black-Box Modeling of Audio Effects
TLDR
Different state-of-the-art deep learning models based on convolutional and recurrent neural networks, feedforward WaveNet architectures and a new model based on the combination of the aforementioned 10 models are analysed and the performance of these models when modeling various analog audio effects is explored.
Modeling Plate and Spring Reverberation Using A DSP-Informed Deep Neural Network
TLDR
A signal processing-informed deep learning architecture is proposed for the modeling of artificial reverberators based on digital reverberators that use sparse FIR filters and the capabilities of deep neural networks to learn such highly nonlinear electromechanical responses are explored.
Differentiable Signal Processing With Black-Box Audio Effects
TLDR
A data-driven approach to automate audio signal processing by incorporating stateful third-party, audio effects as layers within a deep neural network, which can enable new automatic audio effects tasks, and can yield results comparable to a specialized, state-of-the-art commercial solution for music mastering.
InverSynth: Deep Estimation of Synthesizer Parameter Configurations From Audio Signals
TLDR
InverSynth is an automatic method for synthesizer parameters tuning to match a given input sound that is based on strided convolutional neural networks and is capable of inferring the synthesizers parameters configuration from the input spectrogram and even from the raw audio.
SignalTrain: Profiling Audio Compressors with Deep Neural Networks
TLDR
A data-driven approach for predicting the behavior of a given non-linear audio signal processing effect (henceforth "audio effect") using a deep auto-encoder model that is conditioned on both time-domain samples and the control parameters of the target audio effect.
Perceptual loss function for neural modeling of audio systems
  • Alec Wright, V. Välimäki
  • Computer Science
    ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • 2020
TLDR
Listening test results indicate that the use of an A-weighting pre-emphasis filter offers the best improvement among the tested filters, and the proposed perceptual loss function improves the sound quality of neural network models in audio processing without affecting the computational cost.
Lightweight and Interpretable Neural Modeling of an Audio Distortion Effect Using Hyperconditioned Differentiable Biquads
TLDR
This work extends trainable infinite impulse response (IIR) filters to the hyperconditioned case, in which a transformation is learned to directly map external parameters of the distortion effect to its internal filter and gain parameters, along with activations necessary to ensure filter stability.
Neural Modeling of Phaser and Flanging Effects
TLDR
An improved process for accurately measuring the frequency response of a time-varying system over time is introduced, which is used to annotate the neural network training data with the LFO of the effect being modeled.
...
1
2
3
...

References

SHOWING 1-10 OF 28 REFERENCES
Volterra Series and State Transformation for Real-Time Simulations of Audio Circuits Including Saturations: Application to the Moog Ladder Filter
  • T. Hélie
  • Mathematics
    IEEE Transactions on Audio, Speech, and Language Processing
  • 2010
TLDR
The case of the Moog ladder filter is investigated and methods to increase the validity range and to improve the efficiency of Volterra series expansions are detailed on a single stage of the filter.
WaveNet: A Generative Model for Raw Audio
TLDR
WaveNet, a deep neural network for generating raw audio waveforms, is introduced; it is shown that it can be efficiently trained on data with tens of thousands of samples per second of audio, and can be employed as a discriminative model, returning promising results for phoneme recognition.
A Vacuum-Tube Guitar Amplifier Model Using Long/Short-Term Memory Networks
TLDR
This paper reports on the current experiments evaluating the use of long/short-term memory (LSTM) units for modeling the dynamic nonlinear characteristics of a vacuum-tube guitar amplifier and its effects as applied to an electric guitar signal.
A vacuum-tube guitar amplifier model using a recurrent neural network
TLDR
Early results of experiments in using a neural network of the recurrent variety, specifically a Nonlinear AutoRegressive eXogenous (NARX) network, to capture the nonlinear, dynamic characteristics of vacuum-tube amplifiers.
EMULATION OF NOT-LINEAR, TIME-VARIANT DEVICES BY THE CONVOLUTION TECHNIQUE
TLDR
Two methods for obtaining not-linear convolution were pioneered by the authors: Impulse Response switching and diagonal Volterra multiple convolution are described and an evaluation of their performances are provided both in case of memoryless notlinear devices and in cases of devices which show significant memory effects.
Digital implementation of musical distortion circuits by analysis and simulation
TLDR
This research is concerned with preserving the sound of classic musical electronics, namely guitar amplifiers and distortion circuits, through modeling the circuits and emulating their sonic characteristics using efficient techniques to simulate audio circuits.
The Fender Bassman 5F6-A Family of Preamplifier Circuits—A Wave Digital Filter Case Study
TLDR
A Wave Digital Filter study of the preamplifier circuit of 5F6-Abased amplifiers, utilizing recent theoretical advances to enable the simultaneous simulation of its four nonlinear vacuum tube triodes.
A Wavenet for Speech Denoising
TLDR
The proposed model adaptation retains Wavenet's powerful acoustic modeling capabilities, while significantly reducing its time-complexity by eliminating its autoregressive nature.
Deep Voice: Real-time Neural Text-to-Speech
TLDR
Deep Voice lays the groundwork for truly end-to-end neural speech synthesis and shows that inference with the system can be performed faster than real time and describes optimized WaveNet inference kernels on both CPU and GPU that achieve up to 400x speedups over existing implementations.
Wave Digital Simulation of a Vacuum-Tube Amplifier
  • M. Karjalainen, J. Pakarinen
  • Physics, Engineering
    2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings
  • 2006
TLDR
Wave digital filters (WDFs) can be applied to efficient real-time simulation of vacuum-tube amplifier stages, typical in professional guitar amplifiers, which pose nonlinear behavior for desired distortion effects.
...
1
2
3
...