TY - GEN A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Barth, Martin A1 - Tschöpe, Constanze A1 - Schubert, Frank A1 - Wolff, Matthias T1 - Autoencoder-based Ultrasonic NDT of Adhesive Bonds T2 - IEEE SENSORS 2021, Conference Proceedings, Oct 31- Nov 4, Sydney, Australia N2 - We present an approach for ultrasonic non-destructive testing of adhesive bonding employing unsupervised machine learning with autoencoders.The models are trained exclusively on the features derived from pulse-echo ultrasonic signals on a specimen with good adhesive bonding and tested on another specimen with artificially added defects.The resulting pseudo-probabilities indicating anomalies are visualized and presented along to the C-scan of the same specimen. As a result, we achieved improved representation of the defects, allowing their automatic and reliable detection. Y1 - 2021 SN - 978-1-7281-9501-8 U6 - https://doi.org/10.1109/SENSORS47087.2021.9639864 PB - IEEE ER - TY - GEN A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Tschöpe, Constanze A1 - Schubert, Frank A1 - Wolff, Matthias T1 - Paper Tissue Softness Rating by Acoustic Emission Analysis T2 - Applied Sciences N2 - Softness is one of the essential properties of hygiene tissue products. Reliably measuring it is of utmost importance to ensure the balance between customer expectations and cost-effective tissue production. This study presents a method for assessing softness by analyzing acoustic emissions produced while tearing a tissue specimen. The aim was to train neural network models using the corrected results of human panel tests as the ground truth labels and to predict the tissue softness in two- and three-class recognition tasks. We also investigate the possibility of predicting some production parameters related to the softness property. The results proved that tissue softness and production parameters could be reliably estimated only by the tearing noise. KW - acoustic emission KW - machine learning KW - tissue softness analysis Y1 - 2022 U6 - https://doi.org/10.3390/app13031670 SN - 2076-3417 VL - 13 IS - 3 ER - TY - CHAP A1 - Tschöpe, Constanze A1 - Wolff, Matthias A1 - Strecha, Guntram A1 - Duckhorn, Frank A1 - Feher, Thomas A1 - Hoffmann, Rüdiger T1 - Automatisierte Weichheitsprüfung von Papier T2 - ZfP in Forschung, Entwicklung und Anwendung, Graz, 17. - 19. September 2012, DACH-Jahrestagung 2012 Zerstörungsfreie Materialprüfung Y1 - 2012 SN - 978-394-02834-4-3 PB - DGZfP CY - Berlin ER - TY - RPRT A1 - Duckhorn, Frank A1 - Wolff, Matthias A1 - Tschöpe, Constanze T1 - Hidden Markov Model training using Finite State Machines Y1 - 2010 UR - http://www.tu-cottbus.de/fakultaet3/de/kommunikationstechnik/publikationen/reports.html UR - https://www-docs.b-tu.de/fg-kommunikationstechnik/btuauth/_Publikationen/2010_hmmtrn.pdf PB - Technische Universität, Institut für Akustik und Sprachkommunikation CY - Dresden ER - TY - CHAP A1 - Tschöpe, Constanze A1 - Joneit, Dieter A1 - Duckhorn, Frank A1 - Strecha, Guntram A1 - Hoffmann, Rüdiger A1 - Wolff, Matthias T1 - Voice control for measurement devices T2 - AIA-DAGA 2013, proceedings of the International Conference on Acoustics , 18 - 21 March 2013 in Merano Y1 - 2013 PB - DEGA CY - Berlin ER - TY - CHAP A1 - Tschöpe, Constanze A1 - Duckhorn, Frank A1 - Blüthgen, Peter A1 - Richter, Christian A1 - Papsdorf, Gunther A1 - Wolff, Matthias T1 - Miniaturisiertes System zur intelligenten Signalverarbeitung T2 - DGZFP-Jahrestagung 2017, Zerstörungsfreie Materialprüfung, Koblenz, 22.-24.05.2017, Kurzfassungen der Vorträge und Posterbeiträge Y1 - 2017 UR - https://www.google.de/url?sa=t&rct=j&q=&esrc=s&source=web&cd=5&ved=0ahUKEwiOtpCM7-zUAhXBZ1AKHXMyBQYQFghDMAQ&url=https%3A%2F%2Fjahrestagung.dgzfp.de%2FPortals%2F151%2Fdownloads%2FKufa_alles_oP%2520-%2520GESAMT%2520-%2520ohne%2520Anzeigen.pdf%3Fver%3D2017-05-19-135955-633&usg=AFQjCNGFKZug7Xf1FFhVCqafxEFIyDEIeA&cad=rja SP - S. 192 PB - DGZIP CY - Berlin ER - TY - RPRT A1 - Tschöpe, Constanze A1 - Duckhorn, Frank A1 - Wolff, Matthias T1 - Akustische Mustererkennung: Qualitätskontrolle - Vorausschauende Instandhaltung - Zustandsüberwachung Y1 - 2017 UR - https://www.ikts.fraunhofer.de/content/dam/ikts/forschung/energiesysteme_bio_medizintechnik/bio_nanotechnologie/akustische_diagnostik/IKTS_White_Paper_Akustische_Mustererkennung.pdf PB - Fraunhofer-IKTS CY - Dresden ER - TY - GEN A1 - Tschöpe, Constanze A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Wolff, Matthias T1 - Sprachtechnologie und akustische Mustererkennung in der medizinischen Anwendung T2 - 16. Landeskonferenz „Digitalisierung im Gesundheitswesen“ 2021 Y1 - 2021 UR - https://www.digital-agentur.de/veranstaltungen/telemed UR - https://www.digital-agentur.de/fileadmin/06_Bilddatenbank/Gesundheit/Telemed/Praesentationen/Tscho__pe_Impuls_Telemed21.pdf ER - TY - GEN A1 - Wunderlich, Christian A1 - Tschöpe, Constanze A1 - Duckhorn, Frank T1 - Advanced methods in NDE using machine learning approaches T2 - 44th Annual Review of Progress in Quantitative Nondestructive Evaluation, Provo, Utah, USA, 16-21 July 2017 N2 - Machine learning (ML) methods and algorithms have been applied recently with great success in quality control and predictive maintenance. Its goal to build new and/or leverage existing algorithms to learn from training data and give accurate predictions, or to find patterns, particularly with new and unseen similar data, fits perfectly to Non-Destructive Evaluation. The advantages of ML in NDE are obvious in such tasks as pattern recognition in acoustic signals or automated processing of images from X-ray, Ultrasonics or optical methods. Fraunhofer IKTS is using machine learning algorithms in acoustic signal analysis. The approach had been applied to such a variety of tasks in quality assessment. The principal approach is based on acoustic signal processing with a primary and secondary analysis step followed by a cognitive system to create model data. Already in the second analysis steps unsupervised learning algorithms as principal component analysis are used to simplify data structures. In the cognitive part of the software further unsupervised and supervised learning algorithms will be trained. Later the sensor signals from unknown samples can be recognized and classified automatically by the algorithms trained before. Recently the IKTS team was able to transfer the software for signal processing and pattern recognition to a small printed circuit board (PCB). Still, algorithms will be trained on an ordinary PC; however, trained algorithms run on the Digital Signal Processor and the FPGA chip. The identical approach will be used for pattern recognition in image analysis of OCT pictures. Some key requirements have to be fulfilled, however. A sufficiently large set of training data, a high signal-to-noise ratio, and an optimized and exact fixation of components are required. The automated testing can be done subsequently by the machine. By integrating the test data of many components along the value chain further optimization including lifetime and durability prediction based on big data becomes possible, even if components are used in different versions or configurations. This is the promise behind German Industry 4.0. Y1 - 2018 U6 - https://doi.org/10.1063/1.5031519 SN - 0094-243X VL - 37 SP - 1 EP - 8 PB - AIP Publishing CY - College Park, Maryland ER - TY - CHAP A1 - Strecha, Guntram A1 - Wolff, Matthias A1 - Duckhorn, Frank A1 - Wittenberg, Sören A1 - Tschöpe, Constanze T1 - The HMM synthesis algorithm of an embedded unified speech recognizer and synthesizer T2 - Proceedings of the Annual Conference of the International Speech Communication Association 2009, Interspeech 2009, 6 - 10 September, 2009, Brighton, UK Y1 - 2009 SP - 1763 EP - 1766 PB - ISCA CY - Brighton ER - TY - GEN A1 - Schmidt, Ralph Rudi A1 - Hildebrand, Jorg A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Tschöpe, Constanze T1 - A study for laser additive manufacturing quality and material classification using machine learning T2 - 2022 IEEE sensors N2 - This paper demonstrates the use of acoustic emissions (AEs) to monitor the quality, and material used, for the laser additive manufacturing (LAM) process with steel and copper wire. Layers of deposited material (steel or copper) were created using LAM. The quality of these layers was either good or unstable. The AEs were recorded using three sensors, one microphone, and two structure-borne sound probes. The recorded signals were processed and transformed using the fast Fourier method. Then models were trained with the processed data and evaluated using a fivefold cross-validation. Results show that it is possible to accurately classify the materials used during LAM (up to a balanced accuracy [BAcc] score of 0.99). Also, the process quality could be classified with a BAcc score of up to 0.81. Overall, the results are promising, but further research and data collection are necessary for a proper validation of our results. KW - Machine learning KW - Additive manufacturing KW - Neural network KW - Quality monitoring KW - Signal processing KW - Artificial intelligence KW - Data analysis Y1 - 2022 SN - 978-1-6654-8464-0 U6 - https://doi.org/10.1109/SENSORS52175.2022.9967311 SN - 2168-9229 SP - 1 EP - 4 PB - Institute of Electrical and Electronics Engineers (IEEE) CY - Piscataway, New Jersey ER - TY - GEN A1 - Opitz, Joerg A1 - Wunderlich, Christian A1 - Bendjus, B. A1 - Cikalova, U. A1 - Wolf, C. A1 - Naumann, S. A1 - Lehmann, A. A1 - Barth, M. A1 - Duckhorn, Frank A1 - Köhler, B. A1 - Tschöke, K. A1 - Windisch, T. A1 - Tschöpe, Constanze A1 - Moritz, T. A1 - Scheithauer, U. T1 - Materialdiagnose und integrale Prüfverfahren für keramische Bauteile T2 - Keramische Zeitschrift N2 - Hochleistungskeramiken findet man heute häufig als kritische Komponente in neuentwickelten Systemen für Zukunftsanwendungen. Die Zuverlässigkeit des gesamten Systems basiert hierbei oft auf der kritischen keramischen Komponente. Für diese oft neuentwickelten keramischen Materialien werden neue Methoden für die Prozesssteuerung, Materialdiagnostik und Strukturüberwachung benötigt. In diesem Artikel werden drei für die Keramikcharakterisierung am Fraunhofer-Institut für Keramische Technologien und Systeme IKTS weiter entwickelte Technologien und Verfahren beschrieben und deren Einsatz anhand von Beispielen illustriert. Dazu werden die Laser-Speckle-Photometrie, die optische Kohärenztomographie und die Klanganalyse in Kombination mit einer entsprechenden akustischen Mustererkennung als leistungsfähige Verfahren für die Materialdiagnostik im Bereich der keramischen Materialien vorgestellt. T2 - Materials diagnostics and integrated testing technology for ceramic parts Y1 - 2016 U6 - https://doi.org/10.1007/BF03400267 SN - 0023-0561 VL - 68 IS - 4-5 SP - 249 EP - 254 PB - Springer Science and Business Media LLC CY - Berlin ; Heidelberg ER - TY - GEN A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Sobe, Daniel A1 - Tschöpe, Constanze A1 - Wolff, Matthias ED - Grawunder, Sven T1 - Speech-to-text in upper sorbian : current state T2 - Elektronische Sprachsignalverarbeitung 2025 : Tagungsband der 36. Konferenz Halle/Saale, 5.–7. März 2025 N2 - This study presents recent advancements in Upper Sorbian Speech-to-Text (STT) technology. We provide an overview of the Sorbian languages, the available speech and language resources, and the development of an STT system based on a traditional approach, which includes acoustic, pronunciation, and language modeling. Due to the scarcity of resources for Sorbian languages, our approach leverages sub-word and word-class modeling techniques. The word-class modeling is based on Finite-State Transducer definitions, which are applicable to both offline text parsing and integration into the decoding graph of the STT system. Word-class parsing is performed on the speech corpus and utilized for language modeling with complete words, sub-word units, or both. Additionally, the same definitions can be applied to Named Entity Recognition during the post-processing of recognized transcriptions. This approach significantly reduces out-of-vocabulary words and enables greater customization of the recognizer for domain-specific applications. The system was implemented for the real-time transcription of church sermon broadcasts in Upper Sorbian. The domain-specific system achieved performance comparable to fine-tuned OpenAI Whisper models developed also by other initiatives while also providing a resource-efficient solution with semantically tagged recognition results. Y1 - 2025 UR - https://www.essv.de/pdf/2025_109_116.pdf SN - 978-3-95908-803-9 SN - 0940-6832 SP - 109 EP - 116 PB - TUDpress CY - Dresden ER - TY - GEN A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Sobe, Daniel A1 - Tschöpe, Constanze A1 - Wolff, Matthias ED - Karpov, Alexey ED - Delic, Vlado T1 - Preserving Language Heritage Through Speech Technology: The Case of Upper Sorbian T2 - Speech and Computer, SPECOM 2024, Belgrade, Serbia, 25-28 November 2024 N2 - The modern world is facing a crisis with the rapid disappearance of endangered languages, which poses a serious threat to global cultural diversity. Speech Technologies and Artificial Intelligence present promising opportunities to address this crisis by supporting the documentation, revitalization, and everyday use of these vulnerable languages. However, despite recent and remarkable advancements in speech technology, significant challenges persist, particularly for languages with very limited resources and unique linguistic features. This paper details the development of Upper Sorbian speech technologies, focusing on the creation of a practical Speech-to-Text (STT) system as a versatile tool for language preservation. The study explores the current state of Sorbian languages and underscores collaborative efforts with the Foundation for the Sorbian People. Through a series of pilot and successive projects, each phase has contributed to the steady advancement of speech recognition modules and supporting tools, improving their performance, effectiveness and practical usability. KW - Endangered languages, Speech recognition, Upper Sorbian Y1 - 2024 UR - https://link.springer.com/chapter/10.1007/978-3-031-77961-9_1 SN - 978-3-031-77960-2 SN - 978-3-031-77961-9 U6 - https://doi.org/10.1007/978-3-031-77961-9_1 SP - 3 EP - 22 PB - Springer Nature Switzerland, Cham ER - TY - GEN A1 - Leithoff, Ruben A1 - Dilger, Nikolas A1 - Duckhorn, Frank A1 - Blume, Stefan A1 - Lembcke, Dario A1 - Tschöpe, Constanze A1 - Herrmann, Christoph A1 - Dröder, Klaus T1 - Inline monitoring of battery electrode lamination processes based on acoustic measurements T2 - Batteries N2 - Due to the energy transition and the growth of electromobility, the demand for lithium-ion batteries has increased in recent years. Great demands are being placed on the quality of battery cells and their electrochemical properties. Therefore, the understanding of interactions between products and processes and the implementation of quality management measures are essential factors that requires inline capable process monitoring. In battery cell lamination processes, a typical problem source of quality issues can be seen in missing or misaligned components (anodes, cathodes and separators). An automatic detection of missing or misaligned components, however, has not been established thus far. In this study, acoustic measurements to detect components in battery cell lamination were applied. Although the use of acoustic measurement methods for process monitoring has already proven its usefulness in various fields of application, it has not yet been applied to battery cell production. While laminating battery electrodes and separators, acoustic emissions were recorded. Signal analysis and machine learning techniques were used to acoustically distinguish the individual components that have been processed. This way, the detection of components with a balanced accuracy of up to 83% was possible, proving the feasibility of the concept as an inline capable monitoring system. Y1 - 2021 U6 - https://doi.org/10.3390/batteries7010019 SN - 2313-0105 VL - 7 IS - 1 SP - 1 EP - 21 PB - MDPI AG CY - Basel ER - TY - GEN A1 - Lieske, Uwe A1 - Duckhorn, Frank A1 - Tschöpe, Constanze A1 - Harig, Paul A1 - Pallmer, Matthias A1 - Röder, Olaf A1 - Röder, Elisabeth T1 - Verfahren zur akustischen Detektion von Schadinsekten bei der Getreide- und Saatgutlagerung T2 - DGZfP-Jahrestagung Zerstörungsfreie Materialprüfung 2017 : Kurzfassungen der Vorträge und Posterbeiträge : 22.-24. Mai 2017, Koblenz N2 - Getreide bildet eine der wichtigsten Grundlagen für die menschliche und tierische Ernährung. Wie jedes andere pflanzliche Produkt, so ist jedoch Getreide auch Nahrungsquelle für zahlreiche Insekten, die während des Pflanzenwachstums und der Lagerung das Korn befallen können. Jährlich werden etwa 10 % der weltweiten Getreideernte durch Schadinsekten vernichtet. In der Praxis wird der Befall mit Schadinsekten im Rahmen von Stichproben durch visuelle Inspektion, Sieben oder Aufschwemmen der Getreidekörner dokumentiert. Es werden auch einfache Handgeräte mit Mikrofon und Verstärker eingesetzt, um Schädlingsbefall durch Fressgeräusche manuell zu erkennen. Eine frühzeitige Diagnose des Auftretens von Vorratsschädlingen ist für die Einleitung von Gegenmaßnahmen und die Begrenzung des Schadens von großer Bedeutung. Das hier vorgestellte Verfahren bietet die Möglichkeit, die auftretenden Geräusche automatisiert zu analysieren, und ermöglicht so eine schnelle und zuverlässige Erkennung von Schadinsekten. Im Rahmen der Untersuchungen wurden die auftretenden Geräusche von Schadinsekten in den verschiedenen Entwicklungsstadien analysiert und eine geeignete Signalerfassungshardware für den Feldeinsatz entwickelt. Mit den implementierten Signalverarbeitungsalgorithmen der akustischen Mustererkennung erfolgt die automatisierte Auswertung und Klassifizierung der Schädlingsgeräusche. Das Verfahren, die Hardware und die Messergebnisse werden vorgestellt. Y1 - 2017 UR - https://www.ndt.net/article/dgzfp2017/papers/p52.pdf PB - Deutsche Gesellschaft für Zerstörungsfreie Prüfung e.V. CY - Berlin ER - TY - CHAP A1 - Tschöpe, Constanze A1 - Joneit, Dieter A1 - Duckhorn, Frank A1 - Hoffmann, Rüdiger A1 - Strecha, Guntram A1 - Wolff, Matthias T1 - Sprachsteuerung für Mess- und Prüfgeräte T2 - DGZfP-Jahrestagung 2011 Zerstörungsfreie Materialprüfung, 30. Mai - 1. Juni 2011, Bremen, Berichtsband Y1 - 2011 SN - 978-3-940283-33-7 PB - DGZfP CY - Berlin ER - TY - CHAP A1 - Tschöpe, Constanze A1 - Wolff, Matthias A1 - Duckhorn, Frank T1 - Zustandsüberwachung von Magnetventilen anhand der Schaltgeräusche T2 - ZfP in Forschung, Entwicklung und Anwendung, Potsdam, 26. - 28. Mai 2014, DGZfP-Jahrestagung 2014 Y1 - 2014 SN - 978-3-940283-61-0 PB - DGZfP CY - Berlin ER - TY - GEN A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Ju, Yong Chul A1 - Tschöpe, Constanze A1 - Richter, Christian A1 - Wolff, Matthias T1 - Acoustic Resonance Recognition of Coins T2 - 2020 IEEE International Instrumentation and Measurement Technology Conference (I2MTC), 25-28 May 2020, Dubrovnik, Croatia N2 - In this study, we compare different machine learning approaches applied to acoustic resonance recognition of coins. Euro-cents and Euro-coins were classified by the sound emerging when throwing the coins onto a hard surface.The used dataset is a representative example of a small data which was collected in carefully prepared experiments.Due to the small number of coin specimens and the count of the collected observations, it was interesting to see whether deep learning methods can achieve similarly or maybe even better classification performances compared with more traditional methods.The results of the multi-class prediction of coin denominations are presented and compared in terms of balanced accuracy and Matthews Correlation Coefficient metrics. The feature analysis methods combined with the employed classifiers achieved acceptable results, despite the relatively small dataset. Y1 - 2020 SN - 978-1-7281-4460-3 SN - 978-1-7281-4461-0 U6 - https://doi.org/10.1109/I2MTC43012.2020.9129256 PB - IEEE Xplore ER - TY - GEN A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Ju, Yong Chul A1 - Tschöpe, Constanze A1 - Wolff, Matthias ED - Maglogiannis, Ilias ED - Iliadis, Lazaros S. ED - Pimenidis, Elias T1 - Acoustic Resonance Testing of Glass IV Bottles T2 - Artificial Intelligence Applications and Innovations : 16th IFIP WG 12.5 International Conference, AIAI 2020, Neos Marmaras, Greece, June 5–7, 2020, Proceedings, Part II N2 - In this paper, acoustic resonance testing on glass intravenous (IV) bottles is presented. Different machine learning methods were applied to distinguish acoustic observations of bottles with defects from the intact ones. Due to the very limited amount of available specimens, the question arises whether the deep learning methods can achieve similar or even better detection performance compared with traditional methods. Y1 - 2020 SN - 978-3-030-49186-4 SN - 978-3-030-49185-7 U6 - https://doi.org/10.1007/978-3-030-49186-4_17 SN - 1868-4238 SN - 1868-422X VL - Cham SP - 195 EP - 206 PB - Springer International Publishing ER - TY - CHAP A1 - Tschöpe, Constanze A1 - Duckhorn, Frank A1 - Huber, Markus A1 - Meyer, Werner A1 - Wolff, Matthias ED - Karpov, Alexey ED - Jokisch, Oliver ED - Potapova, Rodmonga T1 - A Cognitive User Interface for a Multi-Modal Human-Machine Interaction T2 - Speech and computer : 20th International Conference, SPECOM 2018, Leipzig, Germany, September 18-22, 2018, proceedings Y1 - 2018 UR - https://link.springer.com/chapter/10.1007/978-3-319-99579-3_72 SN - 978-3-319-99578-6 U6 - https://doi.org/10.1007/978-3-319-99579-3 SP - 707 EP - 717 PB - Springer International Publishing CY - Cham ER - TY - GEN A1 - Duckhorn, Frank A1 - Huber, Markus A1 - Meyer, Werner A1 - Jokisch, Oliver A1 - Tschöpe, Constanze A1 - Wolff, Matthias ED - Lacerda, Francisco T1 - Towards an Autarkic Embedded Cognitive User Interface T2 - Proceedings Interspeech 2017, 20-24 August 2017, Stockholm N2 - ucuikt2015 Y1 - 2017 UR - http://www.isca-speech.org/archive/Interspeech_2017/ U6 - https://doi.org/10.21437/Interspeech.2017 SP - 3435 EP - 3436 PB - ISCA ER - TY - CHAP A1 - Tschöpe, Constanze A1 - Duckhorn, Frank A1 - Richter, Christian A1 - Blüthgen, Peter A1 - Wolff, Matthias T1 - Intelligent Signal Processing on a Miniaturized Hardware Module T2 - IEEE SENSORS Proceedings, Glasgow, Scotland, UK, Oct. 29 - Nov. 1, 2017 Y1 - 2017 UR - http://ieeexplore.ieee.org/document/8234023/ SN - 978-1-5090-1012-7 U6 - https://doi.org/10.1109/ICSENS.2017.8234023 N1 - IEEE Catalog Numer: CFP17SEN-ART PB - IEEE CY - Piscataway, NJ ER - TY - CHAP A1 - Tschöpe, Constanze A1 - Duckhorn, Frank A1 - Richter, Christian A1 - Blüthgen, Peter A1 - Wolff, Matthias T1 - An embedded system for acoustic pattern recognition T2 - IEEE SENSORS Proceedings, Glasgow, Scotland, UK, Oct. 29 - Nov. 1, 2017 Y1 - 2017 SN - 978-1-5090-1012-7 U6 - https://doi.org/10.1109/ICSENS.2017.8234380 N1 - IEEE Catalog Numer: CFP17SEN-ART PB - IEEE CY - Piscataway, NJ ER - TY - GEN A1 - Kraljevski, Ivan A1 - Rjelka, Marek A1 - Duckhorn, Frank A1 - Tschöpe, Constanze A1 - Wolff, Matthias ED - Hillmann, Stefan ED - Weiss, Benjamin ED - Michael, Thilo ED - Möller, Sebastian T1 - Cross-Lingual Acoustic Modeling in Upper Sorbian – Preliminary Study T2 - Elektronische Sprachsignalverarbeitung 2021 : Tagungsband der 32. Konferenz Berlin, 3.-5. März 2021 N2 - In this paper, we present a preliminary study for acoustic modeling in Upper Sorbian, where a model of German was used in cross-lingual transfer learning. At first, we define the grapheme and phoneme inventories and map the target phonemes from the most similar German source equivalents. Phonetically balanced sentences for the recording prompts were selected from a combination of general and domain-specific textual data. The speech corpora with a total duration of around 11 hours was collected in controlled recording sessions involving an equal number of females, males, and children. The baseline acoustic model was employed to force-align the speech corpora given the knowledge-based phoneme mappings. How well the mappings were, was evaluated by the phoneme confusions in free-phoneme recognition. The new derived data-driven model with a reduced phoneme set was included in the adaptation and evaluation along with the baseline acoustic model. The model adaptation performance was cross-validated with the “Leave One Group Out” strategy. We observed major improvements in phoneme error rates after adaptation for the knowledge-based and data-driven phoneme mappings. The study confirmed the feasibility of transfer learning for acoustic model adaptation in the case of Upper Sorbian, at the same time demonstrating practical usability with a small vocabulary speech recognition application (Smart Lamp). Y1 - 2021 UR - https://publica.fraunhofer.de/dokumente/N-633297.html SN - 978-3-959082-27-3 SN - 0940-6832 SP - 43 EP - 50 PB - TUDpress CY - Dresden ER - TY - GEN A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Tschöpe, Constanze A1 - Wolff, Matthias T1 - Machine Learning for Anomaly Assessment in Sensor Networks for NDT in Aerospace T2 - IEEE Sensors Journal N2 - We investigated and compared various algorithms in machine learning for anomaly assessment with different feature analyses on ultrasonic signals recorded by sensor networks. The following methods were used and compared in anomaly detection modeling: hidden Markov models (HMM), support vector machines (SVM), isolation forest (IF), and reconstruction autoencoders (AEC). They were trained exclusively on sensor signals of the intact state of structures commonly used in various industries, like aerospace and automotive. The signals obtained on artificially introduced damage states were used for performance evaluation. Anomaly assessment was evaluated and compared using various classifiers and feature analysis methods. We introduced novel methodologies for two processes. The first was the dataset preparation with anomalies. The second was the detection and damage severity assessment utilizing the intact object state exclusively. The experiments proved that robust anomaly detection is practically feasible. We were able to train accurate classifiers which had a considerable safety margin. Precise quantitative analysis of damage severity will also be possible when calibration data become available during exploitation or by using expert knowledge. KW - Machine learning KW - Non-destructive testing KW - Ultrasonic transducers Y1 - 2021 UR - https://ieeexplore.ieee.org/document/9366491 U6 - https://doi.org/10.1109/JSEN.2021.3062941 SN - 1558-1748 VL - 21 IS - 9 SP - 11000 EP - 11008 ER - TY - GEN A1 - Kraljevski, Ivan A1 - Bissiri, Maria Paola A1 - Duckhorn, Frank A1 - Tschöpe, Constanze A1 - Wolff, Matthias T1 - Glottal Stops in Upper Sorbian: A Data-Driven Approach T2 - Proc. Interspeech 2021, 30 August – 3 September, 2021, Brno, Czechia N2 - We present a data-driven approach for the quantitative analysis of glottal stops before word-initial vowels in Upper Sorbian, a West Slavic minority language spoken in Germany. Glottal stops are word-boundary markers and their detection can improve the performance of automatic speech recognition and speech synthesis systems. We employed cross-language transfer using an acoustic model in German to develop a forced-alignment method for the phonetic segmentation of a read-speech corpus in Upper Sorbian. The missing phonemic units were created by combining the existing phoneme models. In the forced-alignment procedure, the glottal stops were considered optional in front of word-initial vowels. To investigate the influence of speaker type (males, females, and children) and vowel on the occurrence of glottal stops, binomial regression analysis with a generalized linear mixed model was performed. Results show that children glottalize word-initial vowels more frequently than adults, and that glottal stop occurrences are influenced by vowel quality. Y1 - 2021 UR - https://www.isca-speech.org/archive/interspeech_2021/kraljevski21_interspeech.html U6 - https://doi.org/10.21437/Interspeech.2021-1101 SP - 1001 EP - 1005 ER - TY - GEN A1 - Kraljevski, Ivan A1 - Duckhorn, Frank A1 - Tschöpe, Constanze A1 - Wolff, Matthias T1 - Convolutional Autoencoders for Health Indicators Extraction in Piezoelectric Sensors T2 - 2020 IEEE Sensors, 25-28 Oct. 2020, Rotterdam, Netherlands, N2 - We present a method for extracting health indicators from piezoelectric sensors applied in the case of microfluidic valves. Convolutional autoencoders were used to train a model on the normal operating conditions and tested on signals of different valves. The results of the model performance evaluation, as well as, the qualitative presentation of the indicator plots for each tested component, showed that the used approach is capable of detecting features that correspond to increasing component degradation. The extracted health indicators are the prerequisite and input for reliable remaining useful life prediction. Y1 - 2020 UR - https://ieeexplore.ieee.org/document/9323023 SN - 978-1-7281-6801-2 U6 - https://doi.org/10.1109/SENSORS47125.2020.9323023 SP - 1 EP - 4 CY - Rotterdam, Netherlands ER - TY - GEN A1 - Maier, Isidor Konrad A1 - Kuhn, Johannes Ferdinand Joachim A1 - Duckhorn, Frank A1 - Kraljevski, Ivan A1 - Sobe, Daniel A1 - Wolff, Matthias A1 - Tschöpe, Constanze T1 - Word Class Based Language Modeling: A Case of Upper Sorbian T2 - Proceedings of The Workshop on Resources and Technologies for Indigenous, Endangered and Lesser-resourced Languages in Eurasia within the 13th Language Resources and Evaluation Conference, Marseille, France N2 - In this paper we show how word class based language modeling can support the integration of a small language in modern applications of speech technology. The methods described in this paper can be applied for any language. We demonstrate the methods on Upper Sorbian. The word classes model the semantic expressions of numerals, date and time of day. The implementation of the created grammars was realized in the form of finite-state-transducers (FSTs) and minimalists grammars (MGs). We practically demonstrate the usage of the FSTs in a simple smart-home speech application, that is able to set wake-up alarms and appointments expressed in a variety of spontaneous and natural sentences. While the created MGs are not integrated in an application for practical use yet, they provide evidence that MGs could potentially work more efficient than FSTs in built-on applications. In particular, MGs can work with a significantly smaller lexicon size, since their more complex structure lets them generate more expressions with less items, while still avoiding wrong expressions. KW - word classes, minimalist grammar, language modeling, speech recognition, Upper Sorbian Y1 - 2022 UR - http://www.lrec-conf.org/proceedings/lrec2022/workshops/EURALI/pdf/2022.eurali-1.5.pdf SN - 978-2-493814-07-4 SP - 28 EP - 35 PB - European Language Resources Association ER - TY - GEN A1 - Maier, Isidor Konrad A1 - Kuhn, Johannes A1 - Duckhorn, Frank A1 - Kraljevski, Ivan A1 - Sobe, Daniel A1 - Wolff, Matthias A1 - Tschöpe, Constanze T1 - Word Class Based Language Modeling: A Case of Upper Sorbian, LREC 2022, Marseille T2 - Language Resources and Evaluation Conference 2022 (LREC 2022), Marseille, 2022-06-13 Y1 - 2022 U6 - https://doi.org/10.5281/zenodo.7501145 ER -