About Me

I am Frank Zalkow, a Senior Scientist at the Fraunhofer Institute for Integrated Circuits IIS, where I work on research and development of text-to-speech synthesis models. I received my Ph.D. (Dr.-Ing.) in 2021 from Friedrich–Alexander University Erlangen–Nürnberg (FAU) for work in semantic music processing and content-based music retrieval. I am interested in deep learning for music and speech, with a focus on applied research and deployment, taking ideas from early prototypes to innovative products.
Education
I earned a Bachelor’s degree in Music Informatics and Musicology in 2012 from the University of Music Karlsruhe, followed by a Master’s degree in Music Informatics in 2015 at the same institution. In 2021, I completed my Ph.D. (Dr.-Ing.) at FAU Erlangen–Nürnberg with the thesis “Learning Audio Representations for Cross-Version Retrieval of Western Classical Music.”
-
Doctoral Thesis: Learning Audio Representations for Cross-Version Retrieval of Western Classical Music (Download, 8.1 MB), Friedrich–Alexander University Erlangen–Nürnberg, 2021.
-
Master Thesis: Automated musical style analysis – Computational exploration of the bass guitar play of Jaco Pastorius on symbolic level (Download, 3.6 MB), University of Music Karlsruhe, 2015.
Career
From 2008 to 2015, I worked at the Max-Reger-Institute Karlsruhe, including contributions from 2010 to 2015 to the Edition of Reger Works (Hybrid Edition) and homepage design from 2013 to 2015. In 2009, at the Institute for Music and Acoustics, ZKM | Center for Art and Media Karlsruhe, I contributed to the electronic music production for the premiere of Karlheinz Stockhausen’s “Strahlen” (“Rays”) for a percussionist and 10-track tape, and from 2010 to 2013 I worked there on digitization and archival projects.
From the winter term 2013/14 to the summer term 2015, I held lectureships at the Institute of Musicology and Music Informatics at the University of Music Karlsruhe . From October 2015 to July 2016, I was a research fellow at the Institute of Musicology, Saarland University, within the DFG project “Computer-aided analysis of harmonic structures” . From August 2016 to July 2021, I was a research fellow at the International Audio Laboratories Erlangen. I joined Fraunhofer IIS in August 2021 as a post-doctoral researcher and have been a Senior Scientist since December 2022. In this capacity, I was heavily involved in the development and deployment of Allinga TTS, a text-to-speech system offered as a SaaS.