Procedural Engine Sounds Dataset

Procedural Engine Sounds Dataset The Procedural Engine Sounds Dataset is a comprehensive collection of synthetically generated and annotated engine audio samples. This dataset contains procedurally generated high-resolution engine sounds free of confounding noises, with detailed time-aligned annotations, designed for research in audio processing, vehicle acoustics, and synthetic sound generation. Dataset Summary Total: 5,935 files, ~19.01 hours, ~24.47 GB Format: 4-channel WAV files at 48 kHz Organization: 8 distinct sets (4 "Full" + 4 "Large" sets) License: CC BY-NC 4.0 Audio Channel Structure Channels 1-2: Stereo engine sound audio Channel 3: Engine speed (RPM × 0.0001) as continuous audio signal Channel 4: Engine torque (Nm × 0.001) as continuous audio signal Research Applications Audio generation models conditioned on engine operation states Audio classification (RPM/torque prediction from sound) Vehicle acoustics and engine sound pattern analysis Procedural audio synthesis research Data augmentation for automotive audio applications Dataset Organization Extended Sets (A, B, C, D): 3,068 files total, ~9.83 hours, ~12.65 GBStandard Sets (E, F, G, H): 2,867 files total, ~9.18 hours, ~11.82 GB Each set includes both audio files and corresponding metadata (JSON summaries and CSV statistics). Technical Specifications Sample Rate: 48 kHz Channels: 4 (quad-channel WAV) Signal Encoding: RPM values scaled by 0.0001, torque values scaled by 0.001 Source: 100% synthetically generated using procedural audio synthesis Citation @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} } Complete documentation, usage examples, and technical requirements are included in the README.md file within the dataset archive.