The Treble acoustic simulation platform has closed an $18 million Series A-2 round, bringing the Reykjavik-based startup deeper into one of AI’s fastest-expanding infrastructure debates: how do you train a machine to hear, not just see?
The round was led by Paladin Capital Group and closed on 17 September 2026, according to The SaaS News. Existing backers KOMPAS VC, Frumtak Ventures, EIC, and Omega ehf also participated. The company has now raised over $40 million in total, with audioXpress reporting the cumulative figure at approximately €36 million across all rounds.
Treble was founded in 2020 by acoustic engineers Finnur Pind and Jesper Pedersen. Pind studied wave physics at the Technical University of Denmark before co-founding the company; his father, Jörgen Pind, spent his career researching how the human ear distinguishes one syllable from another. The lineage shows in the product.
What the Treble Acoustic Simulation Platform Actually Does
Treble’s core argument is that recorded audio scraped from the internet is an inadequate foundation for training AI models that need to understand sound in the real world. The platform uses physics-based simulation, digital twins, and synthetic data generation to create audio datasets that model different rooms, materials, vehicles, devices, speaker configurations, and ambient noise, according to Paladin Capital Group.
‘Audio AI is really a data challenge, and this is where the most opportunities to enable next-generation models and hardware lie,’ Pind said. ‘To date, pretty much all sound-related AI has been made from recordings and data scraped from the internet. We believe that accurate physics simulation can be an alternative way to create data for sound.’
The platform produces those simulations 100 times faster than existing tools, powered by patented technology, according to Business Insider reporting on the company’s earlier fundraise.
For voice AI developers, Treble offers synthetic data generation for speech enhancement, noise suppression, and model training, alongside evaluation services that test how a model performs across different acoustic conditions. Earlier this year it partnered with Hugging Face to launch the Far Field ASR (FFASR) Leaderboard, described in a Nasdaq press release as the first open, community-driven benchmark to evaluate automatic speech recognition models in far-field conditions.
Hardware makers are the other side of the business. Treble works with headphone and speaker companies for virtual prototyping, helping them understand how a product will sound before a physical unit is built. The same logic applies to smart speakers: the platform can test how well a device picks up commands based on where the speaker sits in a room. Lately, smart glasses and AI wearables have become an active area, with Pind describing hearing enhancement as one of the more compelling near-term applications.
‘I’m really excited about the next generation of these devices like headphones and smart glasses that can enable [a feature like] superhuman hearing,’ he said. ‘Maybe you are in a restaurant, and you only want to hear people within two meters of range, or you are in a seminar, and want to mute people around you.’
Why Paladin Is Betting on Sound Infrastructure
The $18 million will fund a push into the US market and an extension of the platform into physical AI, according to the company’s announcement. That means robotics, automotive, and drone companies, where accurate sound interpretation is a function, not a feature.
The bet looks timely. Physical AI systems, including humanoid robots, raised $8.7 billion in disclosed equity between August 2025 and July 2026, with the majority directed toward vision and manipulation rather than sound, according to Tech Funding News. The same outlet projects the voice AI agent market to grow from $2.4 billion in 2024 to $47.5 billion by 2034.
Paladin’s thesis, as articulated by VP Francois Ruether, is that Treble occupies a shared infrastructure layer that individual companies can build on without surrendering proprietary work. ‘Our thesis is that, as more products depend on understanding sound, this infrastructure becomes increasingly valuable across voice AI, wearables, robotics, and physical AI,’ Ruether said. ‘Customers retain ownership of their models, products, and development workflows, while benefiting from a shared foundation of a simulation-native acoustic infrastructure layer.’
Paladin describes itself as a multi-stage investor in cyber, AI, and advanced technology. Its argument for Treble specifically rests on the idea that robots and physical AI systems will need to integrate visual and auditory inputs together, and that multimodal capability is what makes them useful and safe in human environments.
Treble already counts Amazon and Logitech as customers. The $12 million Series A in 2024, led by KOMPAS VC, an early-stage firm that has also backed companies including Vizcab and Findable, proved out the model; this extension is meant to scale it. The question for the next 18 months is whether physical AI builders start treating acoustic simulation as a procurement line item rather than an afterthought.
