TY - JOUR A1 - Hahn, Alexander A1 - Klug, Katharina A1 - Riedmüller, Florian T1 - Digital Empathy: Kombinierte Erfassung über Affective Computing und Eye Tracking JF - Digitale Welt Y1 - 2020 U6 - https://doi.org/10.1007/s42354-020-0278-4 SN - 2569-1996 VL - 4 IS - 3 SP - 24 EP - 27 PB - Springer Science and Business Media LLC ER - TY - CHAP A1 - Hummel, S. A1 - Betzold, C. A1 - Kandasamy, K. A1 - Bordin, S. A1 - Harhausen, G. A1 - Dentel, A. T1 - Experimentelle Validierung von Simulationsmodellen an einem Hardware-in-the-Loop Wärmepumpenprüfstand T2 - BauSIM 2020 N2 - Basierend auf einer experimentellen Messkampagne werden zwei Simulationsmodelle mit unterschiedlichen Modellierungstiefen, welche die Anlagenkomponenten einer Sole-Wasser-Wärmepumpe und eines thermischen Energiespeichers abbilden, validiert. Auf Grundlage von Referenzlastprofilen nach VDI 4655 werden Messungen an einem Hardware-in-the-Loop Wärmepumpenprüfstand unter realitätsnahen Betriebssituationen und Randbedingungen durchgeführt. Die Validierungsergebnisse zeigen generell eine hohe Genauigkeit des detaillierten Modells gegenüber den gemessenen Systemgrößen, aber auch dem vereinfachten Modell, auf. So können beispielsweise absolute Fehler im Ladezustand des Speichers von höchstens 7,2 Prozentpunkten und von der mittleren Speichertemperatur von 1,4 °C ermittelt werden. Der Einsatz des detaillierten Modells ist für eine zukünftige Entwicklung von Regelungsstrategien zu bevorzugen N2 - An experimental validation of two different kind of system simulation models, which both imply a brine/water heat pump as well as a thermal energy storage are carried out. Based on daily load profiles in accordance with VDI 4655, measurements on a hardware-in-the-loop heat pump test bench are performed to evaluate the system components under realistic operating conditions. The validation results confirm the high accuracy of the detailed system model compared to both our measured system parameters and the simplified model. An absolute error of the state of charge and the mean storage temperature of up to 7.2 percentage points and 1.4 °C is determined, respectively. The use of a detailed model for any further heat pump control development is recommended. Y1 - 2020 U6 - https://doi.org/10.3217/978-3-85125-786-1-35 SP - 299 EP - 306 ER - TY - CHAP A1 - Betzold, C. A1 - Dentel, A. A1 - Bordin, S. T1 - Entwicklung und Implementierung von Betriebsführungsstrategien in einem Plusenergiegebäude T2 - BauSIM 2020 N2 - Der steigende Einsatz erneuerbarer Energien im Wohnungsbau erfordert intelligente Lösungen, um die lokal erzeugte Energie vor Ort nutzen zu können. Das Ziel dieser Forschungsarbeit ist es, optimierte Betriebsführungsstrategien für modulierende Wärmepumpen (MWP) mit Speichern zu entwickeln und diese in einem Plusenergie-Reihenhauskomplex anzuwenden, um den Eigenverbrauch der Photovoltaikerzeugung (PV) und die Effizienz zu erhöhen. Der Betrieb und der Vergleich zweier optimierter Betriebsführung-sstrategien zeigen, dass beide einen ähnlich hohen PV-Eigenverbrauch erzielen, aber zu unterschiedlichen Arbeitszahlen führen. N2 - The increasing use of renewable energies in residential buildings requires intelligent solutions in order to be able to use the locally generated energy on site. The aim of this research is the development and the real application of optimized control strategies for modulating heat pumps (MHPs) with thermal storage tanks in plus energy terraced houses, built in 2017. The optimized control strategies aim an increase of PV self-consumption and efficiency. The operation and comparison of two optimized control strategies show, that both reach the same PV self-consumption but different seasonal perfomance factors. Y1 - 2020 U6 - https://doi.org/10.3217/978-3-85125-786-1-16 SP - 144 EP - 151 ER - TY - CHAP A1 - Bayerl, Sebastian P. A1 - Hönig, Florian A1 - Reister, Joelle A1 - Riedhammer, Korbinian T1 - Towards automated assessment of stuttering and stuttering therapy N2 - Stuttering is a complex speech disorder that can be identified by repetitions, prolongations of sounds, syllables or words and blocks while speaking. Severity assessment is usually done by a speech therapist. While attempts at automated assessment were made, it is rarely used in therapy. Common methods for the assessment of stuttering severity include percent stuttered syllables (% SS), the average of the three longest stuttering symptoms during a speech task or the recently introduced Speech Efficiency Score (SES). This paper introduces the Speech Control Index (SCI), a new method to evaluate the severity of stuttering. Unlike SES, it can also be used to assess therapy success for fluency shaping. We evaluate both SES and SCI on a new comprehensively labeled dataset containing stuttered German speech of clients prior to, during and after undergoing stuttering therapy. Phone alignments of an automatic speech recognition system are statistically evaluated in relation to their relative position to labeled stuttering events. The results indicate that phone length distributions differ in respect to their position in and around labeled stuttering events. KW - speech and voice disorders · pathological speech · language Y1 - 2020 U6 - https://doi.org/10.48550/arXiv.2006.09222 ER - TY - CHAP A1 - Mai, Annette A1 - Wagner, Bernhard A1 - Streit, Fabian T1 - Elimination of Current Harmonics in Electrical Machines with Iterative Learning Control N2 - The magnitude of current harmonics depends on the design of an electrical machine. By suppressing these harmonics noise can be reduced and efficiency improved. Iterative Learning Control (ILC) has proven effective in reducing harmonics. One of the challenges of working with ILC is operation at varying speeds. Variable speeds are particularly important for applications like automotive drives. The ILC period length changes during the learning process at varying speeds. Due to fixed sample rates, the number of values processed by the ILC varies with motor speed.This paper proposes a method to solve this problem and uses ILC at varying speeds. The ILC used to eliminate the harmonics is based on the inverse system. The usage of a two-dimensional memory array is proposed. This data structure holds rows for specific speeds between which interpolation is performed, enabling the elimination of errors which are periodically cyclic to one electrical rotation. This includes the reduction of the motor current harmonics. To verify the presented method a permanent magnet synchronous motor with distinctive 5th and 7th harmonics is used. In real-time implementations, limitations of memory and computational capacity occur. KW - Iterative Learning Control, Repetitive Control, Current Harmonics, PMSM, Motor Control Y1 - 2020 U6 - https://doi.org/10.1109/EDPC51184.2020.9388177 VL - 2020 ER - TY - CHAP A1 - Vasquez-Correa, J. C. A1 - Bocklet, T. A1 - Orozco-Arroyave, J. R. A1 - Nöth, E. T1 - Comparison of User Models Based on GMM-UBM and I-Vectors for Speech, Handwriting, and Gait Assessment of Parkinson’s Disease Patients T2 - ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) N2 - Parkinson's disease is a neurodegenerative disorder characterized by the presence of different motor impairments. Information from speech, handwriting, and gait signals have been considered to evaluate the neurological state of the patients. On the other hand, user models based on Gaussian mixture models - universal background models (GMMUBM) and i-vectors are considered the state-of-the-art in biometric applications like speaker verification because they are able to model specific speaker traits. This study introduces the use of GMM-UBM and i-vectors to evaluate the neurological state of Parkinson's patients using information from speech, handwriting, and gait. The results show the importance of different feature sets from each type of signal in the assessment of the neurological state of the patients. KW - Parkinson’s disease KW - GMM-UBM KW - ivectors KW - gait analysis KW - handwriting analysis KW - speech analysis Y1 - 2020 SN - 978-1-5090-6631-5 U6 - https://doi.org/10.1109/icassp40776.2020.9054348 SP - 6544 EP - 6548 PB - IEEE ER - TY - JOUR A1 - Schuster, Maria A1 - Arias-Vergara, Tomás A1 - Müller-Hörner, Rainer A1 - Winterholler, Cordula A1 - Bocklet, Tobias T1 - "Verstehen mich mit der Maske eigentlich noch alle?" JF - Geriatrie-Report Y1 - 2020 U6 - https://doi.org/10.1007/s42090-020-0627-3 SN - 1862-5363 VL - 15 IS - 3 SP - 8 EP - 9 PB - Springer Science and Business Media LLC ER - TY - CHAP A1 - Bayerl, Sebastian P. A1 - Frasetto, Tommaso A1 - Jauernig, Patrick A1 - Riedhammer, Korbinian A1 - Sadeghi, Ahmad-Reza A1 - Schneider, Thomas A1 - Stapf, Emmanuel A1 - Weinert, Christian T1 - Offline Model Guard BT - Secure and Private ML on Mobile Devices N2 - Performing machine learning tasks in mobile applications yields a challenging conflict of interest: highly sensitive client information (e.g., speech data) should remain private while also the intellectual property of service providers (e.g., model parameters) must be protected. Cryptographic techniques offer secure solutions for this, but have an unacceptable overhead and moreover require frequent network interaction. In this work, we design a practically efficient hardware-based solution. Specifically, we build Offline Model Guard (OMG) to enable privacy-preserving machine learning on the predominant mobile computing platform ARM - even in offline scenarios. By leveraging a trusted execution environment for strict hardware-enforced isolation from other system components, OMG guarantees privacy of client data, secrecy of provided models, and integrity of processing algorithms. Our prototype implementation on an ARM HiKey 960 development board performs privacy-preserving keyword recognition using TensorFlow Lite for Microcontrollers in real time. KW - TEE, TrustZone, private ML, speech processing Y1 - 2020 U6 - https://doi.org/10.48550/arXiv.2007.02351 ER - TY - CHAP A1 - Łopatka, Kuba A1 - Bocklet, Tobias T1 - State Sequence Pooling Training of Acoustic Models for Keyword Spotting BT - Proceedings of the Annual Conference of the International Speech Communication Association T2 - Interspeech 2020 N2 - We propose a new training method to improve HMM-based keyword spotting. The loss function is based on a score computed with the keyword/filler model from the entire input sequence. It is equivalent to max/attention pooling but is based on prior acoustic knowledge. We also employ a multi-task learning setup by predicting both LVCSR and keyword posteriors. We compare our model to a baseline trained on frame-wise cross entropy, with and without per-class weighting. We employ a low-footprint TDNN for acoustic modeling. The proposed training yields significant and consistent improvement over the baseline in adverse noise conditions. The FRR on cafeteria noise is reduced from 13.07% to 5.28% at 9 dB SNR and from 37.44% to 6.78% at 5 dB SNR. We obtain these results with only 600 unique training keyword samples. The training method is independent of the frontend and acoustic model topology. KW - keyword spotting KW - machine learning KW - speech recognition Y1 - 2020 U6 - https://doi.org/10.21437/interspeech.2020-2722 SN - 2958-1796 SP - 4338 EP - 4342 PB - ISCA CY - ISCA ER - TY - CHAP A1 - Georges, Munir A1 - Huang, Jonathan A1 - Bocklet, Tobias T1 - Compact Speaker Embedding: lrx-Vector BT - Proceedings of the Annual Conference of the International Speech Communication Association T2 - Interspeech 2020 N2 - Deep neural networks (DNN) have recently been widely used in speaker recognition systems, achieving state-of-the-art performance on various benchmarks. The x-vector architecture is especially popular in this research community, due to its excellent performance and manageable computational complexity. In this paper, we present the lrx-vector system, which is the low-rank factorized version of the x-vector embedding network. The primary objective of this topology is to further reduce the memory requirement of the speaker recognition system. We discuss the deployment of knowledge distillation for training the lrx-vector system and compare against low-rank factorization with SVD. On the VOiCES 2019 far-field corpus we were able to reduce the weights by 28% compared to the full-rank x-vector system while keeping the recognition rate constant (1.83% EER). KW - speaker recognition KW - x-vector KW - low power Y1 - 2020 U6 - https://doi.org/10.21437/interspeech.2020-2106 SN - 2958-1796 SP - 3236 EP - 3240 PB - ISCA CY - ISCA ER -