Industry leaders in audio technology and voice AI will join us to discuss how Treble elevates their products and accelerates their research and development efforts.
In the audio industry, many measurements are required for product development, testing, and recordings across different scenarios, and these are often costly and time consuming. The approach is to use Treble software to replace selected measurements, enable deeper investigation of existing products, support the design of new ones, and generate synthetic data to improve audio algorithms. As an initial step, Treble was validated through simulations of rooms with varying sizes and reverberation times, as well as simulations of videobars and speakerphones, which showed very strong agreement with real measurements. Treble is now used to further develop products and audio algorithms, and the presentation therefore covers room and device simulations across scenarios, microphone array design, and synthetic data generation.
Hugging Face will present an overview of their audio ecosystem including education resources leaderboards and Transformer based ASR and TTS models and explain why high quality data is critical for building reliable audio systems. They will highlight the Treble10 collaboration through the public datasets released on Hugging Face and demonstrate how Treble enables training and evaluation in realistic acoustic conditions with reverb noise and movement while new Treble SDK features simplify development workflows through Hugging Face tooling.
ai-coustics will showcase why speech enhancement models require training data that closely matches real world acoustic conditions and how high-quality synthetic data makes it possible to introduce controlled variability in speaker positions noise sources and room characteristics. They will show how this enables systematic scene construction such as placing multiple speakers at different distances from a microphone to train more specialized and robust models.
CEO - Treble Technologies
Dr. Finnur Pind is the CEO and co-founder of Treble, a company revolutionizing sound simulation and acoustic design through cutting-edge technology. With a background in acoustical engineering and years of experience driving innovation, Finnur has been at the forefront of developing tools that enable designers, engineers, and architects to create better-sounding environments and products. Under his leadership, Treble has become a global leader in cloud-based sound solutions, pushing the boundaries of precision and collaboration in the industry.
Senior Product Manager for the Treble SDK - Treble Technologies
Dr. Daniel Gert Nielsen is a specialist in numerical vibro-acoustics, with a PhD focused on loudspeaker modeling and optimization. His expertise spans acoustic simulation for communication devices and synthetic data generation for machine learning applications. With a strong background in numerical methods and audio technology, he plays a key role in shaping advanced acoustic modeling solutions at Treble.
Senior Simulation Specialist - Treble Technologies
Dr. Sarabeth S Mullins is a specialist in room acoustics, with her dissertation focused on the acoustics of Notre-Dame de Paris. Her expertise includes large-scale room acoustic modeling, interactive acoustics for VR/AR, and human–acoustic interactions. With an interdisciplinary research background and experience in simulation workflows, dataset development, and signal processing, she contributes to high-fidelity acoustic modeling at Treble.
Senior Audio Research Engineer - Treble Technologies
Dr. Georg Götz is a leading expert in acoustics and audio signal processing, with a PhD specializing in data-driven room acoustics modeling. His research and work focuses on large-scale data acquisition, multi-exponential sound energy decay, and 6DoF late reverberation rendering, contributing to more accurate and immersive spatial audio simulations. With extensive experience bridging theoretical research and real-world applications, Dr. Götz plays a key role in advancing next-generation acoustic modeling at Treble.
Audio ML Engineer - Hugging Face
Eric Bezzam is an audio ML engineer at Hugging Face. He received his PhD from EPFL, and previously worked at Snips, Sonos, DSP Concepts, and Fraunhofer IDMT. He was one of the main developers of pyroomacoustics.
Head of Research - ai-coustics
Tim Janke is Head of Research at ai-coustics. He holds a Ph.D. in Machine Learning from TU Darmstadt and has published at NeurIPS. With 10+ deep learning papers, he’s an expert in generative audio and Voice AI, driving breakthroughs that are shaping the future of audio technology at ai-coustics.
Simulation Specialist - GN
Dr. Yauheni Belahurau works as a Simulation Specialist at GN in the Audio Technology department. Prior to joining GN, he completed his PhD at the Technical University of Denmark (DTU), at the Centre for Acoustic-Mechanical Microsystems, followed by a postdoctoral fellowship. His research and professional interests include the design of acoustic metamaterials and transducers, computational optimization, and advanced numerical methods for vibroacoustic problems. His work focuses on leveraging simulation-driven approaches to support product development and the optimization of audio systems.
