Abstract
Human operators often diagnose industrial machinery via anomalous sounds. Automated acoustic anomaly detection can lead to reliable maintenance of machinery. However, deep learning-driven anomaly detection methods often require an extensive amount of computational resources which prohibits their deployment in factories. Here we explore a machine-driven design exploration strategy to create OutlierNets, a family of highly compact deep convolutional autoencoder network architectures featuring as few as 686 parameters, model sizes as small as 2.7 KB, and as low as 2.8 million FLOPs, with a detection accuracy matching or exceeding published architectures with as many as 4 million parameters. Furthermore, CPU-accelerated latency experiments show that the OutlierNet architectures can achieve as much as 21x lower latency than published networks.
Introduction
Acoustic Anomaly Detection (AAD) is a cornerstone of "Industry 4.0," allowing factories to monitor machine health by listening for irregular sounds. While deep learning models excel at this, they are typically "heavyweight." Standard architectures like ResNet-18 or even tailored 4-million-parameter autoencoders require significant memory and computational power. This creates a deployment gap: these models often cannot run on the resource-constrained edge devices, such as microcontrollers or low-power CPUs, found on factory floors. Furthermore, for safety-critical applications, detection must happen in real-time; high latency might cause a system to miss the critical window to shut down a failing machine.
OutlierNets address this gap. They are a new family of ultra-compact convolutional autoencoders designed specifically for detecting industrial machinery failures via sound. By using machine-driven design exploration, these models achieve state-of-the-art accuracy with as few as 686 parameters—representing a 5800x reduction in complexity and up to 21x lower latency compared to previous benchmarks.
Methods and Data
The authors move away from manual architecture design, instead employing a machine-driven design exploration strategy called generative synthesis.
- Generative Synthesis: The researchers defined a set of operational constraints (e.g., parameter count under 100,000 and high detection accuracy) and used an automated generator to traverse the architectural search space.
- Macro & Micro-Architecture: The resulting OutlierNets utilize highly efficient building blocks, including depthwise convolutions, pointwise convolutions, and "replicators."
- Autoencoder Logic: Like standard AAD models, these are trained in an unsupervised manner on "normal" sounds. At inference, the model tries to reconstruct the input sound; if the reconstruction error is high, the sound is flagged as an anomaly.
- Input Processing: Audio is converted into Mel-spectrograms and cropped into one-second windows (32x128), ensuring the system only needs a brief snippet of audio to make a decision.
The study evaluates OutlierNets using the MIMII dataset, a standard benchmark for industrial machine sounds, focusing specifically on Fans and Sliders. The training set contains only "normal" recordings, while the test set is a balanced mix of normal and malfunctioning sounds across various Signal-to-Noise Ratios (SNRs).
Key Results and Insights
The machine-driven process didn't just make the models smaller; it introduced high "micro-architecture diversity." The models use different types of layers in unconventional sequences that a human designer might not consider, but which are mathematically optimal for the task. The exploration found that different machines require different complexities. For example, "Slider" machines were a "tougher" task, leading the generator to produce more complex architectures (with dense latent spaces) compared to the "Fan" models.
The authors compared OutlierNets against a state-of-the-art Convolutional Autoencoder (CAE-MCS) which has 4 million parameters.
- Complexity: The smallest OutlierNet has only 686 parameters and a model size of 2.7 KB. This is small enough to fit into the Static RAM (SRAM) of almost any modern microcontroller.
- Accuracy: Despite being orders of magnitude smaller, OutlierNets achieved an average AUC of 88.8% on sliders, matching or exceeding the 4M-parameter baseline.
- Latency:
- On an Intel Core i5, OutlierNets achieved a runtime of 0.366 μs (21x faster than the baseline).
- On an ARM Cortex A-72 (an embedded CPU), the fastest OutlierNet clocked in at 7.2 μs, compared to 78.2 μs for the baseline.
Conclusion
OutlierNets demonstrate that extreme model compression is possible for acoustic anomaly detection without sacrificing performance. By using one-second audio windows, the authors minimize inherent lag, ensuring the system is viable for real-time safety shutdowns. Future work will explore if these compact architectures generalize well to other industrial sounds like valves or pumps and push the boundaries for even lower-power microcontrollers to enable "always-on" acoustic monitoring with minimal battery drain.
Summary updated