Pulse-Train-Resonator: deep learning architecture for engine sound synthesis

Researchers released Pulse-Train-Resonator (PTR), a deep learning architecture for engine sound synthesis that directly models combustion pressure pulses and exhaust propagation through differentiable Karplus-Strong resonators. Trained on the Procedural Engine Sounds Dataset, PTR achieved improved audio reconstruction compared to harmonic-plus-noise baselines using identical network architecture. The open-source repository includes pre-trained models, training and inference scripts, and supports customization of model parameters.

Physics-Informed Neural Engine Sound Synthesis This repository contains the implementation and trained models for the Pulse-Train-Resonator PTR , a deep learning architecture for engine sound synthesis that directly models combustion pressure pulses and exhaust propagation through differentiable synthesis components. Engine sounds originate from sequential combustion pressure pulses rather than sustained harmonic oscillations. While existing neural synthesis methods model the resulting spectral characteristics, PTR directly models the underlying pulse structure: parameterized pressure waves aligned to engine firing patterns, propagated through differentiable Karplus-Strong resonators simulating exhaust acoustics. The model is trained on the Procedural Engine Sounds Dataset and demonstrates consistent generalization across diverse engine configurations, achieving improved audio reconstruction compared to harmonic-plus-noise baselines using identical network architecture. Physics-informed pulse synthesis incorporating thermodynamic pitch modulation and valve-dynamics envelopes Differentiable Karplus-Strong resonators for exhaust system acoustic modeling Firing-order sequencing with per-cylinder parameterization Generalization across engine types despite fixed architectural priors ptr-model/ ├── checkpoints/ Pre-trained model weights │ └── 2025-08-31 models and weights.zip ├── configs/ Base configuration files ├── scripts/ Training and inference scripts │ ├── train.py Training pipeline with CLI │ └── inference.py Model inference with CLI ├── src/ Source code │ ├── audio/ Audio processing utilities │ ├── data/ Data loading and processing │ ├── models/ Model implementations │ │ ├── hpn model.py Harmonic-Plus-Noise baseline │ │ ├── hpn synth.py HPN synthesis modules │ │ ├── ptr model.py Pulse-Train-Resonator model │ │ ├── ptr synth.py PTR synthesis modules │ │ └── model.py Base model architecture │ ├── training/ Training utilities │ └── utils/ General utilities ├── pyproject.toml Python dependencies and metadata ├── README.md This file └── LICENSE License file - Clone the repository: git clone https://github.com/rdoerfler/ptr-model.git cd ptr-model - Install dependencies: pip install . - Extract pre-trained models: cd checkpoints unzip 2025-08-31 models and weights.zip Download the Procedural Engine Sounds Dataset and put it next to this repository: ├── ptr-model/ This repository ├── dataset/ Procedural Engines Dataset │ ├── A full set Data subsets │ ├── B full set │ ├── C full set │ └── ... Train the HPN variant: python scripts/train.py --model type hpn --dataset C full set Train the PTR variant: python scripts/train.py --model type ptr --dataset C full set Customize model architecture: python scripts/train.py --model type ptr --num harmonics 128 --hidden size 512 --gru size 1024 Generate engine sounds using trained models: python scripts/inference.py Note: As of now, the folder name of checkpoint for inference has to be defined directly within the inference.py CLI support will be inplemented soon. The training and inference scripts use a unified CLI with the following key parameters: --model type : Choose between hpn Harmonic-Plus-Noise or ptr Pulse-Train-Resonator --dataset : Specify dataset name default: C full set --num harmonics : Number of harmonics default: 100 --num noisebands : Number of noise bands default: 256 --hidden size : Hidden layer size default: 256 --gru size : GRU layer size default: 512 Configuration is managed through a base config system combined with CLI parameter overrides. A Harmonic-Plus-Noise HPN baseline is included for comparative evaluation. HPN employs direct harmonic synthesis with systematic inharmonicity modeling and temporally structured noise components, using identical encoder-decoder architecture to PTR. PTR consistently outperforms the HPN baseline across three engine configurations, achieving 5.7% improvement in total validation loss and 21% improvement in audio reconstruction on unseen data. | Dataset | HPN Harm. | HPN STFT | HPN Total | PTR Harm. | PTR STFT | PTR Total | |---|---|---|---|---|---|---| | A | 0.107 | 1.781 | 0.944 | 0.090 | 1.649 | 0.872 | | B | 0.059 | 1.824 | 0.943 | 0.055 | 1.754 | 0.907 | | C | 0.166 | 2.093 | 1.132 | 0.117 | 2.017 | 1.069 | | Mean | 0.111 | 1.899 | 1.006 | 0.088 | 1.807 | 0.949 | This work utilizes the Procedural Engine Sounds Dataset , a comprehensive collection of procedurally generated engine audio with time-aligned control annotations. - 19 hours of engine audio across varied operating conditions - Time-aligned RPM, torque, throttle, and DFCO annotations - Multiple engine configurations and acoustic scenarios - Systematic coverage of engine operating parameters Dataset Availability: Zenodo : https://doi.org/10.5281/zenodo.16883336 https://doi.org/10.5281/zenodo.16883336 Hugging Face Datasets : https://huggingface.co/datasets/rdoerfler/procedural-engine-sounds https://huggingface.co/datasets/rdoerfler/procedural-engine-sounds Evaluation reveals complementary strengths between synthesis approaches: PTR : 5.7% superior validation performance, consistent training-validation transfer HPN : Greater flexibility across engine configurations, robust to harmonic irregularities- Both variants successfully capture authentic engine acoustic behaviors with distinct signatures If you use the Procedural Engines Dataset in your research, please cite: @dataset{doerfler 2025 procedural engine sounds, author = {Doerfler, Robin}, title = {Procedural Engine Sounds Dataset}, month = {August}, year = 2025, publisher = {Zenodo}, version = {1.0}, doi = {10.5281/zenodo.16883336}, url = {https://doi.org/10.5281/zenodo.16883336} } This project is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License CC BY-NC 4.0 . See the LICENSE file for details. You are free to: - Share — copy and redistribute the material in any medium or format - Adapt — remix, transform, and build upon the material Under the following terms: - Attribution — You must give appropriate credit and indicate if changes were made - NonCommercial — You may not use the material for commercial purposes For commercial use, please contact the author. Audio Examples : Supplementary audio examples demonstrating model outputs are available at: https://rdoerfler.github.io/ptr-model-page/ https://rdoerfler.github.io/ptr-model-page/ Dataset : Procedural Engine Sounds Dataset on Zenodo https://doi.org/10.5281/zenodo.16883336 and Hugging Face https://huggingface.co/datasets/rdoerfler/procedural-engine-sounds This research demonstrates systematic integration of physics-informed inductive biases into differentiable synthesis architectures, providing a methodological framework applicable to physically-constrained audio generation beyond automotive contexts. For questions or collaboration opportunities, please contact me or open an issue on this repository. Keywords : Engine Sound Synthesis, Differentiable Signal Processing, Physics-Informed Neural Networks, Inductive Biases, Neural Audio Synthesis