Industrial Sound Analysis – Research and Practice

Working with us

For many years, we have been doing research on audio signal processing and analysis. Additionally, we developed machine learning methods for automated sound analysis, considering technical privacy and data security.

However, cutting-edge research alone cannot succeed without the experience of those who face unsolvable challenges in production every day and can clearly name those. From the users’ point of view, our motivation lies in asking about and understanding the needs of our research partners and customers and responding to them with customized solutions.

In the following, you will find our research activities, through which we have laid the foundation for AI-based acoustic monitoring in production in recent years.

We offer services in the field of research and development and point out numerous possible applications of acoustic monitoring (amo) to our customers in the context of an industrial project. In addition, we work in partly publicly funded research and development projects to develop application-specific solutions together with wide-ranging expert consortia.

ways of working with Fraunhofer IDMT; Industry project, public funding, Fraunhofer networks
Possibilities of joint project work

Current research projects

 

Research project

ML-S-LeAF

Development of machine learning algorithms based on virtual sound data for lightweight construction for quality assurance in additive manufacturing.

 

Research project

AKoS

Acoustic inspection of weld seams of safety-critical components as part of quality assurance

 

Research project

SEC-Learn

Sensor Edge Cloud for federated learning

Research project

RapidKI

Research project

ExtrudEAR

Research project

eLas+

 

Innovationsforum und Leistungszentrum

Netzwerke

IMAMF Innovationsform zum Austausch über KI-basierte, akustische Qualitätssicherung für Fertigungsprozesse

 

Leistungszentrum InSignA – Intelligente Signalanalyse- und Assistenzsysteme

 

Innovationsforum und Leistungszentrum

Networks (German)

IMAMF Innovationsform zum Austausch über KI-basiertes, akustisches Monitoring von Fertigungsprozessen

 

Leistungszentrum InSignA – Intelligente Signalanalyse- und Assistenzsysteme

 

 

Leistungszentrum

InSignA

Showing that it works! Using prototypical demonstrators from different disciplines, we have proven the functionality and effectiveness of novel methods for industrial sound analysis.

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 many German industrial and trade companies for the operation of machines and plants. 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. To detect such leaks, they rely, among other things, on ultrasound testing methods and on 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 leakage detection. First results show that this is generally possible.

Automatic detection and differentiation of electric engines

The engine demo shows the performance of acoustic monitoring in industrial applications. The detection of three different sounding electric engines was chosen as an application example.

In this experiment, three different operating states were simulated: running, broken, and running under heavy load. Each motor is in one of these operational states which sound differently. The researchers placed a MEMS (micro-electro-mechanical systems for signal recording) microphone in front of the engines, which recorded the operating sounds of the running engines. Recordings were made for each engine, which were analyzed using machine learning techniques with and without background noise.

Automatic detection of different materials based on pling sounds

The principle demonstrator "Air-Hockey table" offers visitors to trade fairs and conferences a sporty variety in addition to a business talk.

Using an Air-Hockey table modified for research purposes, the researchers are working on novel approaches for acoustic quality assurance in an industrial context. They use pucks, which are made of different materials and cause different, but very characteristic "pling" noises as soon as they hit the board of the game 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 to make a reliable statement about the material the pucks are made of.

This non-contact method, in which Fraunhofer IDMT combines its many years of expertise in the fields of acoustic measurement technology, signal processing and machine learning, can be used, for example, to detect material defects or in the in-line monitoring of welding processes. If acoustically perceptible defects are already detected during the process, it can be aborted and restarted promptly. Fraunhofer IDMT's acoustic testing method is also non-destructive and thus serves to reduce expensive test scrap.

Automatic detection of material defects based on rolling sounds

The marble track demonstrator illustrates the performance of the acoustic monitoring method in industrial applications. The detection of defects in surfaces was chosen as an application example.

Three differently coated balls roll in random order through the ball track system. The moving noises of the balls are picked up by small microphones and analyzed by machine learning algorithms. In this way, the balls are monitored in real time and identified by their specific rolling sound. The analysis result - the order of arrival - is immediately shown graphically on a display. With this principle demonstrator, the institute shows new acoustic methods for quality assurance - contactless, non-destructive and safely integrated.

Automated acoustic crack detection

Reliably detect hairline cracks in turned parts with airborne sound analysis and AI

As part of the quality inspection, the turned parts shown usually undergo a 100 percent visual final inspection. They are checked for finest hairline cracks, which are hardly visible, but indicate a quality defect. In an experiment, Fraunhofer IDMT investigated to what extent hairline cracks can be detected automatically by airborne sound analysis with AI. As a testing step in the final inspection, the impact of a turned part on a plate was recorded and assigned to the classes "ok" and "not ok" accordingly. The scientists investigated whether the sound spectra of the impact allow a distinction between good and bad parts, which parameters influence these sound spectra (e.g. drop height or material of the plate) and which microphone positions are best suited for the recordings as well as for later analysis. First results showed that it is generally possible to distinguish turned parts into two classes based on the impact sound. 

"When sound engineer and nutcracker collide"

Privacy warning

With the click on the play button an external video from www.youtube.com is loaded and started. Your data is possible transferred and stored to third party. Do not start the video if you disagree. Find more about the youtube privacy statement under the following link: https://policies.google.com/privacy

A little seasonal video greeting - because the possible applications of our AI-based methods are more than diverse.

The data sets have been presented in several scientific publications at international conferences and are supposed to serve the scientific community as potential benchmarks for comparative experiments.

Industrial sound 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
  • IDMT-ISA-Compressed-Air
    An audio database for compressed air leak detection
  • IDMT-TRAFFIC
    An audio database for acoustic traffic monitoring
  • IDMT-FL
    DESED-FL and URBAN-FL: Federated learning datasets for sound event detection

Our scientists exchange information within their research community and with stakeholders from industry and present current research results at scientific meetings and conferences. In addition, we regularly publish articles in professional journals that deal with application topics in our research field, including the German speaking magazines "Maschinenmarkt" or "Fertigungstechnik".

Fritsch, Tobias; Bös, Joachim; Grollmisch, Sascha; Gourishetti, Saichand; Hofmann, Peter; Liebetrau, Judith
Intelligentes akustisches Monitoring durch ausgewählte Mikrofonierungskonzepte
Tagungsband, Fortschritte der Akustik - DAGA 2022

 

Gourishetti, Saichand ; Grollmisch, Sascha; Abeßer, Jakob; Liebetrau, Judith:
Potentials and Challenges of AI-based Audio Analysis in Industrial Sound Analysis
Tagungsband, Fortschritte der Akustik - DAGA 2022

 

Grätzel, M.; Other, S.; Stoll, B.; Rohe, M.; Hasieber, M.; Löhn, T.; Hildebrand, J.; Bergmann, J. P. (Fachgebiet Fertigungstechnik, TU Ilmenau); Kátai, A.; Breitbarth, K.; Bös, J. (Fraunhofer IDMT):
Investigation of the directional characteristics of the emitted airborne sound by Friction Stir Welding for online process monitoring
2nd International Conference on Advanced Joining Processes (AJP 2021)

 

Johnson, David S.; Lorenz, Wolfgang; Tänzer, Michael; Mimilakis, Stylianos; Grollmisch, Sascha; Abeßer, Jakob; Lukashevich, Hanna:
DESED-FL and URBAN-FL: Federated LearningDatasets for Sound Event Detection
29th European Signal Processing Conference, EUSIPCO 2021

 

Abeßer, Jakob; Gourishetti, Saichand; Kátai, András; Clauß, Tobias; Sharma, Prachi; Liebetrau, Judith:
IDMT-Traffic: An Open Benchmark Dataset forAcoustic Traffic Monitoring Research
29th European Signal Processing Conference, EUSIPCO 2021

 

Gourishetti, Saichand; Johnson, David; Werner, Sara; Kátai, András; Holstein, Peter (SONOTEC GmbH):
Partial discharge monitoring using deep neural networks with acoustic emission
Inter-Noise 2021 Congress

 

Gourishetti, Saichand; Werner, Sara; Kátai, András; Liebetrau, Judith:
The Sounds of Partial Discharge
Tagungsband, Fortschritte der Akustik - DAGA 2021

 

Holstein, Peter (SONOTEC GmbH); Gourishetti, Saichand (Fraunhofer IDMT); Fuchs, Karsten (TU Ilmenau); Berger, Frank (TU Ilmenau);  Vyas, Daxeshkumar Harikeshbhai (TU Ilmenau/SONOTEC GmbH):
Acoustics of electrical discharges
27th International Congress on Sound and Vibration, 2021

 

Grollmisch, Sascha; Cano, Estefanía:
Improving Semi-Supervised Learning for Audio Classification with FixMatch
Special Issue Machine Learning Applied to Music/Audio Signal Processing, MDPI, 2021

 

Johnson, David; Grollmisch, Sascha:
Techniques Improving the Robustness of Deep Learning Models for Industrial Sound Analysis
28th European Signal Processing Conference EUSIPCO 2020 

 

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

 

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

 

Helbig, Mareike; Clauß, Tobias; Kepplinger, Sara; Lukashevich, Hanna:
Anwendung von (Luft-)Schallanalyse als ein Verfahren der berührungslosen Qualitätssicherung für die vorausschauende Wartung
Tagungsband, Instandhaltungsforum, 22.-23. Mai 2019, Dortmund

 

Liebetrau, Judith; Grollmisch, Sascha; Nowak, Johannes:
Luftschallbasierte Rissdetektion in kleinen Metallteilen
Proceedings, 44. Deutsche Jahrestagung für Akustik (DAGA) 19.- 22. März 2018, München

 

Nowak, Johannes; Grollmisch, Sascha; Cano Estefanía; Lukashevich, Hanna; Liebetrau, Judith:
Störschallunterdrückung bei Luftschallanalysen in industriellen Fertigungsstrecken
Proceedings, 44. Deutsche Jahrestagung für Akustik (DAGA) 19.- 22. März 2018, München

 

Clauß, Tobias; Abeßer, Jakob; Lukashevich, Hanna; Gräfe, Robert; Häuser, Franz; Kühn, Christian; Sporer, Thomas:
StadtLärm – A distributed system for noise level measurement and noise source identification in a smart city environment
Proceedings, 44. Deutsche Jahrestagung für Akustik (DAGA) 19.- 22. März 2018, München

Articles in professional journals

Clauß, Tobias:
Qualität hören – Das Gehör des Technikers für die Produktion 4.0
ZfP-Zeitung 170, 07/2020

 

Kepplinger, Sara; Krüger, Tobias:
Der Puck macht "pling" 
VDI fachmedien »Technische Sicherheit« 09/2019

 

Liebetrau, Judith; Grollmisch, Sascha; Clauß, Tobias:
Akustische Qualitätskontrolle mit künstlicher Intelligenz
Magazin »Mechatronik« 11/2018

 

Liebetrau, Judith; Grollmisch, Sascha; Clauß, Tobias:
Zisch oder Tzisch? Akustische Qualitätskontrolle in der Getränkeherstellung

Magazin »Getränkeindustrie« 11/2017

 

Liebetrau, Judith; Grollmisch, Sascha; Nowak, Johannes:
Hörbare Fehler – Überwachung von Maschinen und Produkten anhand akustischer Signale

Magazin »QZ – Qualität und Zuverlässigkeit« 09/2017

 

Industrial Sound Analysis

 

Interview-Session

Einschalten lohnt sich

Wir nehmen Sie mit auf eine kleine Tour durch unser Institut und beantworten die meistgestellten Fragen zu unserer Forschung.