Pattern Recognition Lab Secures Funding for VHB Online Course on Speech and Voice Processing

Symbolbild zum Artikel. Der Link öffnet das Bild in einer großen Anzeige.

Erlangen, December 16, 2024 – The Pattern Recognition Lab at Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU), in collaboration with Ostbayerische Technische Hochschule Amberg-Weiden (OTH-AW) and Technische Hochschule Nürnberg Georg Simon Ohm (TH-Nürnberg), has been awarded funding by Virtuelle Hochschule Bayern (VHB) to develop a new CLASSIC vhb online course. The course, titled “Fundamentals of Speech and Voice Processing”, will provide students with cutting-edge knowledge and hands-on experience in the rapidly evolving field of Automatic Speech Recognition (ASR) and speech technologies.

The course, led by Prof. Dr.-Ing. habil. Andreas Maier (FAU), Prof. Dr.-Ing. Christian Bergler (OTH-AW), and Prof. Dr.-Ing. Korbinian Riedhammer (TH-Nürnberg), is set to become a cornerstone of advanced education in speech processing, integrating theoretical foundations with practical applications.

Addressing a Growing Demand for Speech Technology Expertise

Voice and speech technologies are at the heart of modern human-computer interaction, driving innovations in industries such as healthcare, customer service, and education. This interdisciplinary course, taught in English, is designed to equip students with a comprehensive understanding of speech processing principles, phonetics, machine learning techniques, and cutting-edge speech modeling methods.

The curriculum will include:

  • Foundational Concepts: Introduction to ASR, phonetics, speech production, and perception.
  • Feature Extraction Techniques: Spectral analysis, Mel-filter banks, Linear Predictive Coding, and wavelet transformations.
  • Machine Learning for Speech Processing: Hidden Markov Models, Gaussian Mixture Models, Support Vector Machines, and neural networks.
  • Advanced Speech Modeling: Phonetic and language modeling, i-Vectors, and Weighted Finite State Transducers.
  • Hands-on Assignments: Practical implementation in Jupyter Notebooks to visualize and experiment with speech processing methods.

A Collaborative Effort Across Leading Institutions

This initiative brings together expertise from three prestigious Bavarian institutions:

  • FAU Erlangen-Nürnberg: Leading the project under the guidance of Prof. Dr.-Ing. habil. Andreas Maier, a renowned expert in pattern recognition.
  • OTH Amberg-Weiden: Contributing insights in artificial intelligence with a focus on deep learning through Prof. Dr.-Ing. Christian Bergler.
  • TH-Nürnberg: Providing expertise in software architecture and sequence learning under Prof. Dr.-Ing. Korbinian Riedhammer.

Building on International Expertise

The course developers bring a wealth of international experience, having worked in leading institutions such as Stanford University, UC Berkeley, and Siemens. This ensures the course meets the highest academic standards and prepares students to tackle global challenges in the field.

A Step Forward for Digital Education

Designed as an online lecture with accompanying exercises, the course will be accessible to students across Bavaria and beyond. The 4 SWS (Semesterwochenstunden) course will provide 5 ECTS credits, offering significant value for students’ academic progress.

By leveraging the strengths of its consortium members and the support of VHB, this course promises to set new benchmarks in online education for speech and voice processing.