listen, detect, act

With us, you use the acoustic fingerprint of events to make intelligent decisions automatically - for your processes and products.

Machine listening for smart decisions

What do mechanical connections that click into place and approaching emergency vehicles have in common?

 

Both cause acoustic signals that provide valuable information. In addition, they represent scenarios that are difficult to monitor using visual or haptic methods. With the help of machine learning and artificial intelligence, we record and interpret acoustic signals and automatically derive smart recommendations for action. This enables optimized processes and innovative human-technology interactions in a wide range of application fields.

In the automotive sector, acoustic event detection provides for innovative functions, such as detecting the siren of approaching emergency vehicles or monitoring the condition of the vehicle. Acoustic event detection recognizes machine damage in industrial or plant engineering applications before it happens, thus avoiding downtime and making maintenance more predictable (predictive maintenance). In addition, noises occurring in many other areas can also be traced to specific events. In quality control, audible damage, for example the tiniest material fractures in computer chip production, is detected. Detectors are also useful in building security and smart city scenarios for reporting the use of spray cans, cries for help, when glass is broken or other alarm signals. There are also many areas of application in medicine and the care sector including, for example recognizing emergencies with care patients (coughing, a fall, etc.).

 

Our services for you:
 

  • Development of application- and customer-specific recognition technologies that recognize individual events and entire scenes, e.g., mechanical connections that click into place or the detection of a wet surface while driving.
  • Embedding signal processing in hardware systems (microprocessor and DSP), sensor networks and complex system architectures - as an uncomplicated retrofit solution in existing infrastructures, directly in your new product and in mobile applications (apps)
  • Conceptual design and implementation of the optimal microphoning and signal pre-processing for your use case
  • Training and customization of developed sensor systems
  • Your door opener for new service models: Use our smart sensor technology so that your customers can concentrate on their own core business - how about »Reliable Machine Performance as a Service«, for example?

We offer you robust acoustic recognition systems that function reliably even under demanding conditions, without an Internet connection, and at a great distance from the microphone. This ensures a particularly robust, secure and reliable application - even with the highest demands on data protection and data security. Contact us!

Insights into research and practice

 

Press Release / 2.4.2020

Not all clicks are the same

Reliable and documented clicking into place of mechanical connections: Fraunhofer IDMT in Oldenburg is developing a system for inline production monitoring.

 

Press Release / 5.2.2020

Teaching tomorrow’s automobiles to hear

Fraunhofer IDMT designs the »ear of the car« as an important component of autonomous driving.

 

Project KI-MUSIK 4.0

Universal AI Sensor Interface for Industry 4.0

This project is dedicated to the next generation of intelligent, increasingly autonomous manufacturing systems.

 

An insight into our research / 9.12.2019

Understanding pumps

In the interview, Danilo Hollosi reports on how critical noises in industrial plants can be detected by condition monitoring systems.

 

Press Release / 27.3.2019

Minimum machine downtime

Hannover Messe 2019: Fraunhofer IDMT from Oldenburg demonstrates a system for acoustic condition monitoring of industrial machines, which is also available “as a service”

 

Acoustic Monitoring in Smart Cities

Insight into Our Research

How can acoustic data be collected in networked buildings and cities and used for applications in the areas of energy efficiency, security and transport?

 

 

 

Flyer Acoustic Event Detection

Teaching computers the complex skill of hearing.

Oldenburg Branch

The objective of the Oldenburg Branch for Hearing, Speech and Audio Technology HSA of the Fraunhofer Institute for Digital Media Technology IDMT is to transfer scientific findings related to hearing perception and man-machine interaction into technological applications.

R&D-Services and Licencing

Please contact us in case you are interested in our expertise and services.