# UD-ASD: A Unified Diffusion Model for Anomalous Sound Detection

arXiv: https://arxiv.org/abs/2607.12576 · HTML: https://mlsanity.com/paper/2607.12576/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Audio, speech and music, one of 14 directions in Diffusion models.
- Builds on 11 of its 26 references that are in the map.
- First posted on arXiv 14 Jul 2026.

## Placed in

- Diffusion models / [Audio, speech and music](https://mlsanity.com/diffusion/audio-speech-and-music/)

## Builds on (most cited first, 11 of 11)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [MIMII DG: Sound Dataset for Malfunctioning Industrial Machine Investigation and Inspection for Domain Generalization Task](https://mlsanity.com/paper/2205.13879/) (2022)
- [Anomalous Sound Detection using Spectral-Temporal Information Fusion](https://mlsanity.com/paper/2201.05510/) (2022)
- [AnoPatch: Towards Better Consistency in Machine Anomalous Sound Detection](https://mlsanity.com/paper/2406.11364/) (2024)
- [Unsupervised Anomaly Detection and Localization of Machine Audio: A GAN-based Approach](https://mlsanity.com/paper/2303.17949/) (2023)
- [Anomalous Sound Detection using Audio Representation with Machine ID based Contrastive Learning Pretraining](https://mlsanity.com/paper/2304.03588/) (2023)
- [Time-weighted Frequency Domain Audio Representation with GMM Estimator for Anomalous Sound Detection](https://mlsanity.com/paper/2305.03328/) (2023)
- [ASD-Diffusion: Anomalous Sound Detection with Diffusion Models](https://mlsanity.com/paper/2409.15957/) (2024)
- [Hierarchical Metadata Information Constrained Self-Supervised Learning for Anomalous Sound Detection Under Domain Shift](https://mlsanity.com/paper/2309.07498/) (2023)

## Abstract (arXiv)

Anomalous Sound Detection (ASD) aims to determine whether faults have occurred by monitoring sounds. Existing methods detect a limited range of anomalies, exhibit poor generalization, or train a separate model for each machine. Diffusion models possess strong generalization and can generate specific data with condition guidance. We propose a unified diffusion model only with a small module. The audio is first transformed into log-Mel spectrograms. The lightweight module embeds machine IDs into condition embeddings, guiding the model to reconstruct data for specific machines. Then diffusion model reconstructs data with condition, using Gaussian Mixture Models to fit the distributions of reconstruction errors. Our unified model could monitor multiple machine types and learn more fundamental feature spaces with cross-domain learning. Experiments on DCASE2022 Challenge Task 2 show that our model achieves 3.44% AUC and 2.52% pAUC improvements over baseline, validating its effectiveness.