Prof. Dr.-Ing. Christian Bergler

Prof. Dr. Christian Bergler

Department of Computer Science
Chair of Computer Science 5 (Pattern Recognition)

Martensstr. 3
91058 Erlangen

Analysis of Underwater Audio Recordings of Marine Animals (Killer Whales)

For marine biologists, the interpretation and understanding of underwater audio recordings is essential. Based on such recordings, possible conclusions about behavior, communication and social interactions of marine animals can be made. Despite a large number of biological studies on the subject of orca vocalizations, it is still difficult to recognize a structure or semantic significance of the orca signals in order to be able to derive any patterns of behavior.

Northern Resident Killer Whale during the fieldwork expedition 2018/2019 in Vancouver Island, British Columbia, Canada (copyright by Anthro-Media)

 

Due to a lack of  techniques and computational tools, hundreds of hours of underwater recordings are still listened to by marine biologists in order to detect potential orca vocalizations. In a post process these identified orca signals will be analyzed and categorized. The main goal is to provide a robust method which is able to automatically detect orca calls within underwater audio recordings.

Orca
An Orca fin and its grey Saddle Patch form the “Face of an Orca” during the fieldwork expedition 2018/2019 in Vancouver Island, British Columbia, Canada (copyright by Anthro-Media)

 

A robust detection of orca signals in connection with the associated situational video recordings and behaviour descriptions (provided by several researchers on site) can provide potential information about communication (kind of a language model) and behaviors (e.g. hunting, socializing). Furthermore, the orca signal detection algorithm can be used in conjunction with a localization software to provide the researchers on the field a more efficient way of searching orca populations. For more information about the project please contact me at christian.bergler@fau.de.

Orca Pod
Orca family (Matriline/Pod) of several motherly related Individuals during the fieldwork expedition 2018/2019 in Vancouver Island, British Columbia, Canada (copyright by Anthro-Media)

 

Television Documentary Research Results

The joint efforts of the entire project team to decode the language of the Orcas will be featured on German television (3Sat). The documentation can be watched under the following link:

https://www.3sat.de/wissen/wissenschaftsdoku/die-sprache-der-wale-102.html

 

“Whale Hello” from a Northern Resident Killer Whale during the fieldwork expedition 2019/2020 in Vancouver Island, British Columbia, Canada (copyright by Prof. Dr.-Ing. Elmar Nöth)

Academic CV

Since 04/2018: PhD Student, Pattern Recognition Lab, Friedrich-Alexander University Erlangen-Nürnberg

  • Research focus on Deep Learning Applied to Animal Linguistics in particular the analysis of underwater audio recordings of marine animals (killer whales)

04/2016 – 03/2018: Software Engineer and external PhD Student (FAU), e.sigma Technology GmbH, Ilmenau

  • Software developer for communication, voice processing, and guidance systems (automated speech recognition, text-to-speech systems, language model adaption)

03/2014 – 03/2016: Master of Engineering in Information Technology and Automation, Ostbayerische Technical University of Applied Science (OTH), Amberg – Weiden, Location: Amberg

  • Focus on information technology
  • Master thesis: Parametrization of various speaker characteristics for the generation of emotions within a High Quality Limited Domain Text – to – Speech – System (HQLD-TTS)

10/2010 – 03/2014: Bachelor of Engineering in Industrial Engineering, Ostbayerische Technical University of Applied Science (OTH), Amberg – Weiden, Location: Weiden

  • Focus on mechanical engineering
  • Bachelor thesis: Development of a freeTTS-based speech-synthesis system for the training of air traffic controllers

09/2009 – 07/2010: Vocational Diploma (Fachabitur), Berufsoberschule (BOS), Schwandorf

 

Teaching Experience

  • since 03/2019: PhD Student, Friedrich-Alexander University Erlangen-Nürnberg (FAU)
    • Exercises: Introduction to Pattern Recognition (IntroPR), Deep Learning
    • Seminar: Automatische Analyse von Stimm-, Sprech- und Sprachstörungen bei Sprachpathologien
    • Praktikum: Representation Learning
  • 10/2011 – 06/2012: Student teaching assistant, Ostbayerische Technical University of Applied Science (OTH), Amberg – Weiden, Location: Weiden
    • Modules: Mathematics/Statistics, Data Processing and Programming, Physics, Industrial Economics

 

Practical Experience

  • 04/2016 – 03/2018: Software Engineer, e.sigma Technology GmbH, Ilmenau
    • Automatic Speech Recognition (ASR) for Air Traffic Control (ATC)
    • Controller Pilot Data Link Communication (CPDLC)
    • Advanced Voice Processing (Text – to – Speech (TTS) Engine, limited Domain TTS Voice Creation)
    • Communication Framework (VoIP Standard)
    • Realtime Audio and Video Streaming
    • Distributed System Communication via Message Broker Systems
  • 08/2013 – 11/2015: Student trainee, e.sigma Technology GmbH, Ilmenau
    • Software development, testing, maintenance, integration and documentation for the java-based freeTTS text-to-speech engine
  • 08/2012 – 01/2013: Student trainee, Continental Automotive GmbH, Regensburg
    • Support programming, integration and verification of Manufacturing Execution Systems (=MES) for all worldwide Continental
      Chassis & Safety plants

Projects

2018

  • Deep Learning Applied to Animal Linguistics

    (FAU Funds)

    Term: April 1, 2018 - April 1, 2022
    Abstract
    Deep Learning Applied to Animal Linguistics in particular the analysis of underwater audio recordings of marine animals (killer whales):

    For marine biologists, the interpretation and understanding of underwater audio recordings is essential. Based on such recordings, possible conclusions about behaviour, communication and social interactions of marine animals can be made. Despite a large number of biological studies on the subject of orca vocalizations, it is still difficult to recognize a structure or semantic/syntactic significance of orca signals in order to be able to derive any language and/or behavioral patterns. Due to a lack of techniques and computational tools, hundreds of hours of underwater recordings are still manually verified by marine biologists in order to detect potential orca vocalizations. In a post process these identified orca signals are analyzed and categorized. One of the main goals is to provide a robust and automatic method which is able to automatically detect orca calls within underwater audio recordings. A robust detection of orca signals is the baseline for any further and deeper analysis. Call type identification and classification based on pre-segmented signals can be used in order to derive semantic and syntactic patterns. In connection with the associated situational video recordings and behaviour descriptions (provided by several researchers on site) can provide potential information about communication (kind of a language model) and behaviors (e.g. hunting, socializing). Furthermore, orca signal detection can be used in conjunction with a localization software in order to provide researchers on the field with a more efficient way of searching the animals as well as individual recognition.

    For more information about the DeepAL project please contact christian.bergler@fau.de.

Publications

2024

Conference Contributions

2023

Journal Articles

2022

Journal Articles

Conference Contributions

2021

Journal Articles

Conference Contributions

2020

Conference Contributions

2019

Journal Articles

Conference Contributions

Lectures

No matching records found.

Supervising Bachelor/Master Theses

Type Title Status
BA thesis Animal-Independent Signal Enhancement Using Deep Learning finished
MA thesis Contextual Meta Knowledge Integrated into a Fully-Automated Multi-Stage Deep Learning Framework for Killer Whale Individual Classification finished
BA thesis Deep Orca Image Denoising Using Machine-Generated Binary Killer Whale Masks finished
MA thesis Killer Whale Sound Source Localization Using Deep Learning finished
BA thesis Web-Based Server-Client Software Framework for Killer Whale Indivirual Recognition finished
BA thesis Automatic Bird Individual Recognition in Multi-Channel Recording Scenarios finished
MA thesis A Robust Intrusive perceptual audio quality assessment based on convolutional neural network finished
BA thesis Start, follow, read, stop: Incorporating new steps into end-to-end full-page handwriting recognition method finished
MA thesis Orca Individual Identification based on Image Classification Using Deep Learning finished
BA thesis Killer Whale Sound Type Generation Using Generative Adversarial Networks (GAN) finished
BA thesis Killer Whale Echolocation Click Detection in Noisy Underwater Recordings Using Deep Learning finished
MA thesis Killer Whale Matriline Classification based on Deep Learning finished
MA thesis Deep Feature Learning and Clustering – A Fully Unsupervised Approach for Identifying Orca Communication Patterns finished
MA thesis Generative Adversarial Networks for Speech Vocoding finished
MA thesis Semi-supervised Feature Learning for Orca Audio Signals using a Convolutional Autoencoder finished