International Journal For Multidisciplinary Research

E-ISSN: 2582-2160   •   Impact Factor: 9.24

A Widely Indexed Open Access Peer Reviewed Multidisciplinary Bi-monthly Scholarly International Journal

Call for Paper Volume 8, Issue 5 (September-October 2026) Submit your research before last 3 days of October to publish your research paper in the issue of September-October.
Submit Research Paper Join as a Reviewer

Plagiarism is checked by the leading plagiarism checker TurnItIn

Call for Paper

Volume 8 Issue 5
September-October 2026

Submit your research paper

Indexing Partners
Academia Advanced Sciences Index Bielefeld Academic Search Engine CiteSeer DRJI Google Scholar Independent Search Engine & Directory Network ISI (International Scientific Indexing) Issuu Mendeley Research Networks
RefSeek ResearcherId - Thomson Reuters ResearchGate Scirus Scribd Semantic Scholar UTeM - Universiti Teknikal Malaysia Melaka Wiki for Call for Papers WorldCat Zenodo

EchoEdge: A Lightweight Deep Learning and IoT Framework for Acoustic Machinery Diagnostics

Author(s) Mr. Aravindhan R, Mr. Pradeep Kumar R, Prof. Charulatha R.T
Country India
Abstract When a fan or pump stops without warning, the bill is rarely just the spare part. Plants still lose hours to that kind of failure. We listen to the machine instead of bolting on extra probes: a cheap microphone is enough to hear wear, imbalance or a leak before the asset seizes. EchoEdge is the working stack we assembled for that job. A small convolutional net reads log-mel spectrograms of industrial audio. A two-mode console sits on top. The whole thing also ships as a Windows folder that runs offline. Audio comes from MIMII, the public industrial-machine set on Zenodo (record 3384388). MIMII already blends machine sound with plant noise at three signal-to-noise ratios: +6 dB, 0 dB and −6 dB. One net is trained on all three mixes and then, for each 10 s clip, says normal or abnormal. On a held-out slice of unit id_00, mixed across those three ratios, we measured 96.71% accuracy, 97.48% precision and 95.08% recall on the abnormal class at P(abnormal) ≥ 0.40. Streamlit drives the console. It can either walk through a live 10 s-per-clip demo, with staged faults and a mix of the three noise packs, or score one uploaded recording. Playback, a live spectrum, a latching silence alarm and the portable folder sit around that core. The numbers are for one unit under MIMII’s controlled packs. They are enough for a lab demo and for operator practice on this unit. They are not a licence to drop the same weights on every machine in a plant.
Keywords Acoustic Anomaly Detection, Industrial Condition Monitoring, Edge AI, Deep Learning, Log-Mel Spectrogram, Internet of Things, Predictive Maintenance
Field Computer > Artificial Intelligence / Simulation / Virtual Reality
Published In Volume 8, Issue 4, July-August 2026
Published On 2026-08-28
DOI https://doi.org/10.36948/ijfmr.2026.v08i04.85903
View / Download PDF File

Share this



E-ISSN 2582-2160
CrossRef

CrossRef DOI prefix of IJFMR is 10.36948/ijfmr

Downloads
Research Paper Format Copyright Permission Form and Undertaking Form

All research papers published on this website are licensed under Creative Commons Attribution-ShareAlike 4.0 International License, and all rights belong to their respective authors/researchers.

CC-BY-SA Open Access

About IJFMR
Fees & Payment
Current Issue
Publication Archive 		Submit Research Paper
Track Submission Status
Publication Guidelines
Publication Ethics
Peer Review & Plagiarism 		Join as a Reviewer
Editors & Reviewers
Reviewer Referral Program
Get Reviewer Membership Certi. 		Website/Journal Policies
Usage Policy
Content Policies
Privacy Policy
Contact Us +91-9687-828-838 editor@ijfmr.com
Powered by Sky Research Publication and Journals