Neuron Soundware
Neuron Soundware is a Czech developer of acoustic condition monitoring systems that use machine learning to detect mechanical faults in industrial and transport equipment, headquartered in Prague.
Neuron Soundware was founded in Prague in 2016 by Pavel Konečný, who leads the company as CEO. Business databases report a team of roughly 10–15 people; the company is backed by investors including Inven Capital and J&T Ventures.
Its technology addresses a recurring gap in the digital systems and software stack of rail operators: components that carry no built-in sensors but reveal their condition through sound.
Products
The core system combines piezoelectric structure-borne sensors and microphones with the nEdge and nBox edge computing units, which process audio data locally rather than streaming it to the cloud. Neural networks trained on machine sound data compare live audio against nominal behaviour and flag developing faults.
Results are visualised in the nGuard web platform, where operators can review alerts and stored sound samples over time. This is acoustic monitoring as one of the AI and machine learning applications in rail: continuous, automated listening where periodic manual inspection was previously the only option.
Market position
In rail and metro environments, the company’s reference work centres on Prague public transport operator DPP, where its system continuously monitors 21 metro escalators for early fault detection. Similar escalator monitoring has been deployed in the Vienna metro.
The company has also applied its technology to rolling stock, including fault detection in pneumatic train door components. Beyond transport, deployments span power generation and manufacturing.
Current activity
According to the company, it was named among the top ten global startups in railroad innovation by StartUs Insights in 2024. It continues to expand its predictive maintenance platform across the power, manufacturing, and transport sectors.
