aiqbench analysis
This research introduces MemNMF, a novel approach for anomalous sound detection in machine condition monitoring that leverages Linear Predictive Coding spectra instead of standard spectrograms. By utilizing an NMF-based memory module with attention weighting, the method demonstrates improved robustness against noise and transients compared to conventional autoencoder baselines. The study validates these gains across multiple datasets and operating conditions, offering a more reliable framework for detecting faults in industrial machinery using only normal training data.
MemNMF: Memory-Augmented NMF on LPC Spectra for Anomalous Sound Detection
arXiv:2607.22086v1 Announce Type: cross Abstract: Autoencoder-based anomalous sound detection is attractive for machine condition monitoring because it can be trained using only normal recordings and yields an interpretable anomaly score from reconstruction error. Most prior work uses spectrogram autoencoders, but reconstructing detailed time--frequency patterns is sensitive to noise and transients, and models can reconstruct some anomalous inputs well, weakening normal--anomaly separation. We propose MemNMF, a constrained reconstruction method that operates on the Linear Predictive Coding spectrum, a compact estimate of the spectral envelope. MemNMF initializes a memory module from an NMF dictionary learned on normal LPC spectra and reconstructs each input as an attention-weighted combination of prototypical normal spectral patterns. Experiments on MIMII and DCASE 2020 Task 2 across multiple machine types and operating conditions show that LPC-spectrum inputs improve a standard autoencoder baseline and that MemNMF yields further gains, with especially strong robustness under noisy, non-stationary settings.
Abstract:Autoencoder-based anomalous sound detection is attractive for machine condition monitoring because it can be trained using only normal recordings and yields an interpretable anomaly score from reconstruction error. Most prior work uses spectrogram autoencoders, but reconstructing detailed time--frequency patterns is sensitive to noise and transients, and models can reconstruct some anomalous inputs well, weakening normal--anomaly separation. We propose MemNMF, a constrained reconstruction method that operates on the Linear Predictive Coding spectrum, a compact estimate of the spectral envelope. MemNMF initializes a memory module from an NMF dictionary learned on normal LPC spectra and reconstructs each input as an attention-weighted combination of prototypical normal spectral patterns. Experiments on MIMII and DCASE 2020 Task 2 across multiple machine types and operating conditions show that LPC-spectrum inputs improve a standard autoencoder baseline and that MemNMF yields further gains, with especially strong robustness under noisy, non-stationary settings.
Sound (cs.SD); Machine Learning (cs.LG)
arXiv:2607.22086 [cs.SD]
(or arXiv:2607.22086v1 [cs.SD] for this version)
https://doi.org/10.48550/arXiv.2607.22086 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 239 words
Related coverage
AnomaMind: Agentic Time Series Anomaly Detection with Tool-Augmented Reasoning
arXiv · Jun 10, 2026
Presentation: Platform Teams Enabling AI - MCP/Multi-Agentic Tools across Linkedin
InfoQ · Jun 5, 2026
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · Sep 4, 2026
MA-DLE: Speech-based Automatic Depression Level Estimation via Memory Augmentation
arXiv · Jun 11, 2026