Research Topics

Synergetic use of expertise - Fraunhofer IDMT draws on many years of expertise and thus generates value for I4.0
Synergetic use of expertise - Fraunhofer IDMT draws on many years of expertise and thus generates value for I4.0

Fraunhofer IDMT relies on a strong network of competence and partners from the fields of education, science and industry to implement projects. In order to realize their existing competencies in industrial applications, the scientists of the IMA business unit are always looking for new use-cases for acoustic condition monitoring or quality control.

Experiment for the automatic detection of compressed air leaks

Reliably and automatically detect audible leaks thanks to airborne sound analysis and machine learning methods.

Compressed air is an indispensable resource for the operation of machines and plants for many German industrial and trade companies. At the same time, it also represents a high cost factor on the electricity bill. On average, companies waste 30 percent of the energy generated due to the unnoticed escape of expensive compressed air. In order to detect such leaks, the methods used include ultrasonic testing and the mere hearing of trained personnel.

Fraunhofer IDMT has now tested in an experiment whether this "hearing" can be automated and reproduced using microphones in combination with machine learning methods in order to develop a reliable system for detecting leaks. First results show that this is generally possible.

More pling, less test scrap

The principle demonstrator "Air Hockey Table" gives visitors to trade fairs and conferences a sporty alternative besides the professional exchange.

Using an air hockey table modified for research use, novel methods for acoustic quality assurance in an industrial context have been developed. The demonstrator uses Pucks, which are made of different materials and cause different, but very characteristic "pling" sounds as soon as they hit the board of the table. During the game, these acoustic signals occur so frequently and irregularly that they can be used for analysis by means of machine learning methods in order to make a reliable statement about the material from which the pucks are made.

Fraunhofer IDMT has many years of expertise in the fields of acoustic metrology, signal processing and machine learning which are all integrated in this non-contact method, which can be used, amongst others, to detect material defects or for in-line monitoring of welding processes. If acoustically perceptible errors are found already in the production process, this process can be aborted and restarted promptly. Fraunhofer IDMT's acoustic test method is also non-destructive and therefore allows the reduction of expensive test scrap.

Intelligent acoustic measurement technology

  • Sound recording (structure-borne sound and airborne sound measurements in disturbance-free and disturbed environments)
  • Customer-specific HW/SW development of multi-microphone arrays including data pre-processing and data transmission
  • Networking of sensors and time-synchronous transmission technology

Signal analysis and processing

  • Single and multi-channel signal processing in connected systems and machinery (directional filtering, source separation, feature extraction, acoustic fingerprinting)

Machine Learning

  • Models of machine learning based on data patterns for the evaluation of unknown data and for the analysis of arbitrary sensor data including microphones

Media-specific and privacy-enhancing security technologies

  • Integrity and authenticity of (media) data and metadata
  • Confidentiality of (media) data and metadata
  • Legally compliant data management
  • Data security-friendly data analysis and thus data sovereignty

Scientific publications


Grollmisch, Sascha (TU Ilmenau); Johnson, David and Liebetrau, Judith (Fraunhofer IDMT): 
Visualizing Neural Network Decisions for Industrial Sound Analysis
Proceedings SMSI 2020 – Sensor and Measurement Science International (conference was cancelled)


Grollmisch, Sascha (TU Ilmenau); Johnson, David; Krüger, Tobias and Liebetrau, Judith (Fraunhofer IDMT): 
Plastic Material Classification using Neural Network based Audio Signal Analysis
Proceedings SMSI 2020 – Sensor and Measurement Science International (conference was cancelled)


Foss, Jeremy; Shirley, Ben; Malheiro, Benedita; Kepplinger, Sara; Nixon, Lyndon; Philipp, Basil; Mezaris, Vasilieos; Ulisses, Alexandre:
DataTV 2019: 1st International Workshop on Data-Driven Personalisation of Television
Proceedings of the 2019 ACM International Conference on Interactive Experiences for TV and Online Video, Salford (Manchester), United Kingdom

Articles in professional journals

Liebetrau, Judith; Grollmisch, Sascha:
Predictive Maintenance: Acoustic Condition Monitoring via airborne sound analysis

Processing Magazine, 2017/09

Published datasets


  • IDMT-ISA-Electric-Engine
    An audio database for the automatic analysis of operational states of electric enginges
  • IDMT-ISA-Metal-Balls
    An audio database for the automatic surface detection of metal balls
  • IDMT-ISA-Tubes
    An audio database for the automatic detection of bulk materials
  • IDMT-ISA-Pucks
    An audio database for the automatic detection of air-hockey pucks of different plastic materials