Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Fine-Grained Engine Fault Sound Event Detection Using Multimodal Signals

Dennis Fedorishin, Livio Forte, Philip Schneider, Srirangaraj Setlur, Venu Govindaraju

TL;DR

This paper introduces a multimodal SED framework that detects fine-grained engine faults of automobile engines using audio and accelerometer-recorded vibration and shows that fusing features from audio and vibration improves overall engine fault SED capabilities.

Abstract

Sound event detection (SED) is an active area of audio research that aims to detect the temporal occurrence of sounds. In this paper, we apply SED to engine fault detection by introducing a multimodal SED framework that detects fine-grained engine faults of automobile engines using audio and accelerometer-recorded vibration. We first introduce the problem of engine fault SED on a dataset collected from a large variety of vehicles with expertly-labeled engine fault sound events. Next, we propose a SED model to temporally detect ten fine-grained engine faults that occur within vehicle engines and further explore a pretraining strategy using a large-scale weakly-labeled engine fault dataset. Through multiple evaluations, we show our proposed framework is able to effectively detect engine fault sound events. Finally, we investigate the interaction and characteristics of each modality and show that fusing features from audio and vibration improves overall engine fault SED capabilities.

Fine-Grained Engine Fault Sound Event Detection Using Multimodal Signals

TL;DR

This paper introduces a multimodal SED framework that detects fine-grained engine faults of automobile engines using audio and accelerometer-recorded vibration and shows that fusing features from audio and vibration improves overall engine fault SED capabilities.

Abstract

Sound event detection (SED) is an active area of audio research that aims to detect the temporal occurrence of sounds. In this paper, we apply SED to engine fault detection by introducing a multimodal SED framework that detects fine-grained engine faults of automobile engines using audio and accelerometer-recorded vibration. We first introduce the problem of engine fault SED on a dataset collected from a large variety of vehicles with expertly-labeled engine fault sound events. Next, we propose a SED model to temporally detect ten fine-grained engine faults that occur within vehicle engines and further explore a pretraining strategy using a large-scale weakly-labeled engine fault dataset. Through multiple evaluations, we show our proposed framework is able to effectively detect engine fault sound events. Finally, we investigate the interaction and characteristics of each modality and show that fusing features from audio and vibration improves overall engine fault SED capabilities.
Paper Structure (12 sections, 2 equations, 3 figures, 3 tables)

This paper contains 12 sections, 2 equations, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Method
  3. Dataset
  4. Model Architecture
  5. Implementation Details
  6. Large-Scale Pretraining
  7. Experiments
  8. Evaluation Metrics
  9. Quantitative Results
  10. Qualitative Results
  11. Conclusion
  12. Acknowledgments

Figures (3)

  • Figure 1: a) Proposed engine fault sound event detection architecture using audio and vibration engine recordings. b) Proposed pretraining strategy using a large-scale weakly-labeled engine fault dataset with our supervised contrastive loss. Best viewed with zoom and color.
  • Figure 2: Example engine fault detections. Events above the red line are ground truth labels and below the red line are predicted events.
  • Figure 3: Example failure cases of detecting engine fault sound events.