Showing papers by "Reinhold Orglmeister published in 2013"

Computing MMSE Estimates and Residual Uncertainty Directly in the Feature Domain of ASR using STFT Domain Speech Distortion Models

Abstract: In this paper we demonstrate how uncertainty propagation allows the computation of minimum mean square error (MMSE) estimates in the feature domain for various feature extraction methods using short-time Fourier transform (STFT) domain distortion models. In addition to this, a measure of estimate reliability is also attained which allows either feature re-estimation or the dynamic compensation of automatic speech recognition (ASR) models. The proposed method transforms the posterior distribution associated to a Wiener filter through the feature extraction using the STFT Uncertainty Propagation formulas. It is also shown that non-linear estimators in the STFT domain like the Ephraim-Malah filters can be seen as special cases of a propagation of the Wiener posterior. The method is illustrated by developing two MMSE-Mel-frequency Cepstral Coefficient (MFCC) estimators and combining them with observation uncertainty techniques. We discuss similarities with other MMSE-MFCC estimators and show how the proposed approach outperforms conventional MMSE estimators in the STFT domain on the AURORA4 robust ASR task.

SASET: a program for series analysis of small‐angle scattering data

Abstract: Technische Universität Berlin, Stranski Laboratorium für Physikalische und Theoretische Chemie, Institut für Chemie, Strasse des 17 Juni 124, Sekretariat TC7, 10623 Berlin, Germany, Technische Universität Berlin, Elektronik und Medizinische Signalverarbeitung, Sekretariat EN3, Einsteinufer 17, 10587 Berlin, Germany, and Helmholtz-Zentrum Berlin, Lise-Meitner Campus, Hahn-MeitnerPlatz, 14109 Berlin, Germany. Correspondence e-mail: michael.gradzielski@tu-berlin.de

Online learning algorithms for principal component analysis applied on single-lead ECGs.

Abstract: This article evaluates several adaptive approaches to solve the principal component analysis (PCA) problem applied on single-lead ECGs. Recent studies have shown that the principal components can indicate morphologically or environmentally induced changes in the ECG signal and can be used to extract other vital information such as respiratory activity. Special interest is focused on the convergence behavior of the selected gradient algorithms, which is a major criterion for the usability of the gained results. As the right choice of learning rates is very data dependant and subject to movement artifacts, a new measurement system was designed, which uses acceleration data to improve the performance of the online algorithms. As the results of PCA seem very promising, we propose to apply a single-channel independent component analysis (SCICA) as a second step, which is investigated in this paper as well.

Schallfeldmodellierung von Ultraschall-Transversalwellen-Prüfköpfen - Modellierung von Prüfkopfarrays mit der Elastodynamischen Finiten Integrationstechnik und der Punktquellensynthese

Abstract: Kurzfassung Ultraschallmessverfahren werden seit Langem erfolgreich fur Prufaufgaben, wie zum Beispiel die Ermittlung von Bauteilgeometrien, im Bauwesen eingesetzt. Um die Schallbundelung der Prufkopfe zu verbessern, werden oft mehrere Prufkopfe in einem Array angeordnet. Fur die Entwicklung der Prufkopfarrays ist es zweckmasig, mithilfe von Modellierungen den Einfluss der Arraygrose, der Anordnung der Einzelprufkopfe im Array und der Pruffrequenz auf das Schallfeld zu untersuchen. Dazu werden in der Praxis erfolgreich Modellierungen auf der Basis der Elastodynamischen Finiten Integrationstechnik (EFIT) und der Punktquellensynthese angewendet. Beim Aufbau der Arrays kommen zurzeit immer haufiger Transversal-Punktkontaktprufkopfe zur Anwendung. Daher wird bei der Beschreibung der Schallfeldberechnung insbesondere auf die der Transversalwellenprufkopfe eingegangen und mithilfe von Modellierungen und Experimenten gezeigt, wie sich die Prufkopfpositionierung auf die Richtcharakteristik auswirkt.

A novel method for motion artifact removal in wearable ppg sensors based on blind source separation

Abstract: The recent development of healthcare systems has provided a significant contribution to ambulatory patient monitoring. In that context, signal quality and disturbances induced by noise or motion artifacts play an important role in the field of signal processing tasks. Especially the Photo- plethysmogram (PPG) is very liable to movement artifacts which severely hamper the extraction of vital parameters like the heart rate or oxygen saturation. To record patient movements, an innovative sensor system is proposed, which acquires accelerometer data next to the PPG. As in Adap- tive Noise Cancelers, we propose to use the acceleration as reference to recover corrupted PPGs by means of the Blind Source Separation. Sophisticatedmethods of ICAhave been used, resulting in a novel approach for artifact suppression in the PPG that has been tested on laboratory datasets.

Automatic Validation and Quality Based Readjustment of Manually Scored EEG Arousal.

Abstract: A knowledge of arousals during sleep is impor- tant to attain a deeper understanding regarding cardiovas- cular diseases. Manual scoring is time consuming and not always accurate. Automatic approaches are even worse inter alia due to inaccurate learning data. This paper presents an algorithm to improve the accuracy of manu- ally scored data. Also a measure of quality is introduced to judge the automatically estimated results.