In the context of a European automotive manufacturer, the use of a piston helium compressor in transmission gear hardening is vital for production. This study addresses the challenge of maintaining uninterrupted operations in the tempering furnace, as equipment failure could lead to production delays and scrap generation. Previously, entire compressors were replaced due to critical incidents. The proposed solution employs IoT devices and non-intrusive sensors to gather and analyze acoustic data. AI and machine learning algorithms assess compressor sounds, promptly detecting deviations from the norm as anomalies. Benefits encompass early detection of potential failures, real-time asset monitoring, preemptive alerts, streamlined inspection prioritization, and reduced costs tied to failures and scrap.