Neuron Soundware Uses AI-Powered Audio Analysis for Industrial Predictive Maintenance

Founded in 2016 by Pavel Konečný through the StartupYard accelerator, Neuron Soundware provides audio-based diagnostics and predictive maintenance solutions for industrial machinery using AI and IoT technologies. The Czech company analyses sounds emitted by engines, compressors, pumps, and production equipment to detect anomalies indicating potential mechanical failures.

The Prague-based company has secured €120,000 in debt financing from Repsol Foundation as it expands its predictive maintenance platform across power generation, manufacturing, and transportation sectors. Neuron Soundware deploys piezoelectric sound sensors and condenser microphones to capture structure-borne and airborne sound from machinery, processing data through edge computing devices or cloud infrastructure.

The system uses machine learning algorithms trained on machine sound data to analyze audio patterns. It operates continuously, monitoring for mechanical wear indicators and generating alerts for maintenance scheduling. Their solution function across temperature ranges and can be deployed on individual machines or production lines. So far the company has received industry awards including recognition at Hannover Messe, DCASE rankings, and railroad innovation listings.

Founder: Pavel Konečný

Sources: Crunchbase, Neuron Soundware, Innovation Origins, EIT Manufacturing