Josef Novak - Chief Innovation Officer

Josef Novak

Chief Innovation Officer

Zurich, Zurich, Switzerland

Join Prog.AI to see contacts

Summary

👤

Senior

🎓

Top School

Josef Novak is a Chief Innovation Officer and co-founder with over 16 years of engineering and 20 years of international IT experience focused on spoken language technologies, speech-to-text and analytics. He led product and solution design for Dialog Composer and Speech Analytics at Spitch, shaping roadmaps and integrating client, partner and engineering requirements into deployable VUX solutions for contact centers and telephony. With a Ph.D. in ASR from The University of Tokyo and research stints at Yandex and NICT, he combines deep WFST-era academic expertise with hands-on ML engineering—currently contributing ASR recipes to the widely used ESPnet toolkit. Multilingual and globally mobile, Josef brings business-level Japanese (N1), lived experience across the US, Japan and Europe, and dual US citizenship plus Swiss permanent residence, enabling seamless collaboration across markets. He pairs founder-level strategy with low-level C++/Python fixes (e.g., Phonetisaurus RNNLM patches), making him equally at home in boardrooms and codebases.

16 years of coding experience

13 years of employment as a software developer

B.A., Linguistics, B.A., Linguistics at UC Santa Barbara

Ph.D., ASR, Ph.D., ASR at 東京大学 / The University of Tokyo

Research Student, ASR, Research Student, ASR at 東京工業大学 / Tokyo Institute of Technology

English, Japanese, German

Github Skills (12)

pytorch10

deeplearning-ai10

speech-recognition10

c-language10

deep-learning10

cprogramming-language10

python10

speech-to-text9

machine-translation9

speech-synthesis8

text-to-speech7

kaldi6

Programming languages (9)

TypeScriptShellC++CJavaScriptGoJupyter NotebookRuby

Github contributions (5)

AdolfVonKleist/Phonetisaurus

Jun 2014 - Oct 2021

Phonetisaurus G2P

Role in this project:

Back-end Developer

Contributions:1 review, 105 commits, 26 PRs in 7 years 4 months

Contributions summary:Josef primarily contributed to the `phonetisaurus` project by modifying and updating server-side components. Their work includes fixing bugs related to numerical errors within the RNNLM implementation, specifically addressing double includes in the `util.o` object. They also worked on server-side configuration, configuration of a twisted server, and updates to client-side components for evaluation. The user's commits demonstrate proficiency in C++ and Python, touching upon aspects of language modeling.

g2pspeech-to-textspeech-recognitionpronunciation-dictionarylexicon

espnet/espnet

Mar 2020 - Dec 2020

End-to-End Speech Processing Toolkit

Role in this project:

ML Engineer

Contributions:2 reviews, 10 commits, 2 PRs in 8 months

Contributions summary:Josef primarily contributes to the development and refinement of the French Polyphone ASR recipe within the ESPnet toolkit. They added a new recipe for the French Polyphone dataset and made subsequent updates, including changes to data preparation scripts and experiment management. The commits involve adjustments to the training and decoding scripts, demonstrating an understanding of the ASR pipeline and its configurations.

speech-recognitionspeech-separationchainerspoken-language-understandingspeech-processing

Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.

Request Free Trial