Personalize your sound environment

Experience the power of deep learning in the very noisy world. Hear only what is important to you by filtering out unwanted sounds and enhancing relevant ones in real-time.

Vision

As a contract research and development partner for Sonova AG, we help advance their mission to improve quality of life with artificial intelligence. We contribute to the development of software that uses deep learning to distinguish speech from noise, filter unwanted sounds in real time, and provide personalized listening experiences. Together with Sonova we support millions of hearing aid users, enhancing their ability to interact socially in challenging environments.

Technology

To create real-time systems that run on mobile devices, we combine cutting-edge algorithms for efficient implementation of deep neural networks with novel deep learning techniques for detecting, filtering, and modifying sound.

Team

Drawing on expertise from a variety of disciplines including deep learning, natural language processing, computational neuroscience, and neuromorphic engineering, our team is at the forefront of groundbreaking research. We are over 20 passionate professionals from more than 10 different countries dedicated to bringing our vision to life.

As a small, supportive team within a global leader, we pair startup agility with Sonova’s scale, audiological expertise, and world-class resources. Everyone at Audatic has the chance to shape how we work, contribute their ideas and have a direct impact on the products we deliver – creating technology that can transform lives. 

Publications

Preuss, M., & Krylova, A. (2025). Spheric speech clarity 2.0. Phonak Insight.

Hasemann, H., & Krylova, A. (2024). Revolutionary speech understanding with spheric speech clarity. Phonak Technical Paper.

Hasemann, H., & Krylova, A. (2024). Artificial intelligence in hearing aid technology. Phonak Technical Paper.

Diehl, P. U., Zilly, H., Sattler, F., Singer, Y., Kepp, K., Berry, M., Hasemann, H., Zippel, M., Kaya, M., Meyer-Rachner, P., Pudszuhn, A., Hofmann, V. M., Vormann, M., & Sprengel, E. (2023). Deep learning-based denoising streamed from mobile phones improves speech-in-noise understanding for hearing aid users. Frontiers.

Diehl, P. U., Singer, Y., Zilly, H., Schönfeld, U., Meyer-Rachner, P., Berry, M., Sprekeler, H., Sprengel, E., Pudszuhn, A., & Hofmann, V. M. (2023). Restoring speech intelligibility for hearing aid users with deep learning. Scientific Reports.

Diehl, P. U., Thorbergsson, L., Singer, Y., Skripniuk, V., Pudszuhn, A., Hofmann, V. M., Sprengel, E., & Meyer-Rachner, P. (2022). Non-intrusive deep learning-based computational speech metrics with high-accuracy across a wide range of acoustic scenes. PLOS One.

Jobs

We are always looking for smart, driven, and entrepreneurial people who are excited about our mission.