About Me
Hi, I'm Bastian, dad, hobby photographer, software developer, and signal processing scientist in Germany.
Both photography and programming are about telling stories. Expressing a complex algorithm in simple terms, or stunning vista in a compelling picture. Finding order and simplicity in a chaotic scene or code base. Therein lies the challenge and joy of life.
You can get in touch with me at bastibe at bastibe dot de.
Education
DOCTOR OF ENGINEERING, Carl von Ossietzky University Oldenburg (2014-2021)
“Pitch of Voiced Speech in the Short‐Time Fourier Transform: Algorithms, Ground
Truths, and Evaluation Methods”, Magna cum Laude
My doctoral work started with designing a pitch estimation algorithm for voiced speech, but I soon realized that there already existed more than enough algorithms. The real challenge of pitch estimation therefore was finding good databases and visualizations and error metrics for evaluating various algorithms. This lead to the development of an audio-specific database, a custom batch processing framework robust to scientific code running amok, and an interface library for running Matlab code from Python, all released as Open Source software.
MASTER OF SCIENCE, Carl von Ossietzky University Oldenburg (2012-2014)
BACHELOR OF ENGINEERING, University of Applied Sciences Oldenburg (2006-2010)
Two parts of Hearing Technology and Audiology, an engineering course about all things audio across medicine, acoustics, electronics, and signal processing. I put a major focus on Digital Signal Processing and Psychoacoustics.
Work
FRAUNHOFER IDMT, Oldenburg (2021-present)
Machine Learning Developer - C, Python, Kaldi
I inherited the core of an automatic speech recognition system, a DNN engine and a WFST decoder. My job was to modernize this legacy code into something maintainable and understandable, and port it to various desktop, server, and embedded platforms.
—
LECTURER, University of Applied Sciences Oldenburg (2013-present)
Introduction to Programming, Data and Algorithms, Applied Programming and various lab classes and thesis supervisor
My lecture started as an advanced course on signal processing methods in Matlab, but I evolved it quickly into a general advanced programming course in Python, data analysis and GUI development.
—
OPEN SOURCE DEVELOPER (2014-present)
Data Analysis, Audio, Image Processing - Python, C, Lua
The open internet has revolutionized the way we work with data. When I started, there were not yet any good Python libraries for working with audio data, so I had to write my own. SoundFile and SoundCard in particular have almost a million downloads per month, and are included in tens of thousands of other projects. Find a more complete list of my contributions here.
—
SENNHEISER AUDIO LABS, Waldshut-Tiengen (2012-2014)
Signal Processing Engineer - Python, C++, Qt, QML
During my Master's studies, I worked part-time on simulating audio algorithms and building a QML based GUI library for audio applications.
—
SOMIX AUDIO LABS, Waldshut-Tiengen (2010-2012)
Embedded Software Developer - C, C++, Lua, Blackfin, Qt, Python
We built the processing/software side of a mixing console for live performances. My job was the user interface, including the GUI and hundreds of buttons and dials and faders. I built a Lua scripting system running on a Blackfin DSP, a bare-metal GUI library with a graphical editor, a custom build system, and various audio effects work.
—
ALGORITHMIX (2010)
Freelance Mac Software Developer - Obj-C, C, OpenGL
I built a GUI Mac application around an existing audio analysis library.
