Björn W. Schuller – författare

This book constitutes the proceedings of the Second International Conference on Signal Processing and Computer Vision, SIPCOV 2025, held in Silchar, India, during August 8–9, 2025. The 36 full papers were included in this were carefully reviewed and selected from 146 submissions. They focus on Signal Processing, Computer Vision, and Biomedical Image Processing.

Die 5. Auflage des Klassikers zur Statistischen Informationstechnik erfährt eine substantielle Erweiterung im Bereich des maschinellen Lernens. Sie bietet somit einen ausgezeichneten Überblick über die beiden wichtigen Themen Mustererkennung/Signalverarbeitung und Maschinelles Lernen.Die Autoren behandeln die Signalerkennung im Rauschen und die Mustererkennung sowie die Parameter- und Signalschätzung. Moderne Verfahren wie Wavelet-Transformation oder Clusterbildung mit unscharfen Partitionen werden berücksichtigt. Neben klassischen Verfahren der Detektion werden neuere, z.B. auf neuronale Netze und kernelbasierten Methoden aufbauende Klassifikatoren diskutiert. Die Parameterschätzung behandelt neben Bayes- und Maximum-Likelihood-Ansätzen auch adaptive Verfahren. Wiener- und Kalman-Filter sind Beispiele zur Signalschätzung. Die Grundlagen werden durch Anwendungsbeispiele aus der Praxis erläutert.Geeignet für Studierende und für Ingenieure in der Praxis.

This book provides the reader with the knowledge necessary for comprehension of the field of Intelligent Audio Analysis. It firstly introduces standard methods and discusses the typical Intelligent Audio Analysis chain going from audio data to audio features to audio recognition.  Further, an introduction to audio source separation, and enhancement and robustness are given. After the introductory parts, the book shows several applications for the three types of audio: speech, music, and general sound. Each task is shortly introduced, followed by a description of the specific data and methods applied, experiments and results, and a conclusion for this specific task. The books provides benchmark results and standardized test-beds for a broader range of audio analysis tasks. The main focus thereby lies on the parallel advancement of realism in audio analysis, as too often today’s results are overly optimistic owing to idealized testing conditions, and it serves to stimulate synergies arising from transfer of methods and leads to a holistic audio analysis.

This book provides the reader with the knowledge necessary for comprehension of the field of Intelligent Audio Analysis. It firstly introduces standard methods and discusses the typical Intelligent Audio Analysis chain going from audio data to audio features to audio recognition.  Further, an introduction to audio source separation, and enhancement and robustness are given. After the introductory parts, the book shows several applications for the three types of audio: speech, music, and general sound. Each task is shortly introduced, followed by a description of the specific data and methods applied, experiments and results, and a conclusion for this specific task. The books provides benchmark results and standardized test-beds for a broader range of audio analysis tasks. The main focus thereby lies on the parallel advancement of realism in audio analysis, as too often today’s results are overly optimistic owing to idealized testing conditions, and it serves to stimulate synergies arising from transfer of methods and leads to a holistic audio analysis.