Robots That Can't Communicate Naturally Won't Be Adopted, Says Treble

Humanoid robots have become a fixture at major tech events, showcasing impressive locomotion and manipulation. Yet, communication and auditory perception remain strikingly underdeveloped, according to Gunnar Pétur Hauksson, co-founder of Treble Technologies, a company focused on audio AI.

Hauksson argues that for robots to operate in the real world, they need more than vision. Real-world environments are loud and chaotic, with background noise and reverberation. Current audio systems in robots tend to break down in such conditions. The robotics industry has prioritized vision because of abundant data and mature simulation tools, but this has left audio behind. Platforms like NVIDIA Isaac Sim have enabled rapid progress in vision, but the environments used for training are mostly silent.

Treble examines this gap through an evolutionary lens. Humans allocate 15-20% of sensory processing to hearing, the second most after vision. This suggests that sound is critical for communication and situational awareness. Without robust auditory perception, robots will struggle to be accepted by humans.

Robots are held to a higher standard in communication than in movement. A robot that mishears or responds poorly becomes frustrating, breaking trust. This is why audio is disproportionately important for adoption. The issue is not just technical performance but the breakdown of trust.

Audio data is difficult to obtain and scale. Unlike visual data, it requires context like spatial relationships and environmental acoustics. Simulation is challenging because sound is governed by wave physics, making accurate modeling computationally demanding. Many current systems rely on simplified models, limiting their generalization.

Treble is working to shift this paradigm by building physically accurate acoustic simulation. The company generates high-fidelity acoustic data for training, with a low sim-to-real gap. This is already used by leading technology companies. For robotics, this represents an inflection point. As vision and locomotion become commoditized, communication will become the differentiator.

Treble is actively seeking partnerships with robotics and physical AI teams that recognize this shift. The company believes that robots that succeed will be defined by how naturally they communicate. The difference between a robot that works and one that is accepted will largely come down to communication, which is fundamentally auditory.

Gunnar Pétur Hauksson is co-founder and chief commercial officer of Treble Technologies. He has over a decade of experience in sales, marketing, and business development. Treble is based in Reykjavík, Iceland, and builds high-fidelity acoustic simulation infrastructure for sound in buildings, devices, vehicles, and intelligent machines.