Sanas, a speech AI company that changes how voices sound on live calls, announced on April 14, 2026 that it had acquired Tomato.ai, a startup that built real-time voice transformation for call centers and communications networks.
The companies did not disclose the price or other terms. Tomato.ai co-founder Ofer Ronen joined Sanas to lead its global telecom strategy, with a focus on expanding across carriers and communications platforms.
According to Outsource Accelerator, the purchase is Sanas's third acquisition in less than two years. Sanas said it now holds eight patents covering real-time speech processing, voice transformation and low-latency inference.
What Tomato.ai built
Tomato.ai developed what Sanas calls zero-shot, real-time voice transformation: software that alters speech as it happens, without first having to train on the individual speaker. Its first commercial product was an accent softening tool for call center agents that runs on a PC and works with Zoom, Teams and Meet, making speech less accented in real time while keeping the speaker's voice recognisable. It launched alongside an AI noise cancellation feature that removes background noise and nearby voices.
Sanas said the main attraction was Tomato.ai's work inside live communications infrastructure. The startup had built production integrations with VoIP environments, communications platforms and systems close to telecom carriers, designing its speech technology to run as part of high-volume call systems rather than as an add-on.
In comments reported by Outsource Accelerator, Ronen said much speech AI is built as a feature added on top, while at carrier and platform level it has to work as part of the system itself.
“Voice sits at the center of how enterprises and platforms operate, yet the technology supporting real-time communication has lagged behind. Bringing Tomato.ai into Sanas accelerates our ability to make speech work reliably and consistently anywhere communication happens at scale.”
Why Sanas is buying
Sanas describes itself as an AI lab building a real-time speech processing layer for enterprises and global communications, and already works with communications platforms and carrier partners, according to Speech Technology magazine. It says it went to market in 2023, has grown to $60 million in annual recurring revenue and is on track to pass $120 million.
The company wants its speech layer to run inside the systems that carry calls rather than on top of them. With Tomato.ai's integrations, Sanas said it plans to extend speech enhancement, accent transformation and language understanding across enterprise environments, carrier networks and communications platforms, as well as latency-sensitive uses in gaming, live streaming and interactive media.
Sharath Keshava Narayana, Sanas's co-founder and chief executive, said voice is central to how enterprises and platforms work, but that the technology behind real-time communication had lagged, and that adding Tomato.ai would speed up Sanas's effort to make speech work reliably wherever communication happens at scale.
The founders and the funding
Ronen is a serial entrepreneur who had founded and sold three companies before Tomato.ai, two of them to Google and one to IAC, according to Sanas. He led Tomato.ai as co-founder and chief executive.
Tomato.ai announced a $2.1 million seed round in June 2024, at the same time as the launch of its accent softening product. Cardumen Capital and Recursive Ventures led the round and Gaingels joined as a new investor. The company said at the time it planned to deploy accent softening with call centers in the Philippines, India and Pakistan. Tomato.ai's total funding and headcount were not disclosed in the deal announcements.
What happens to customers
Sanas did not say whether Tomato.ai's products would continue to be sold under their own name. It described Tomato's technology and operating experience as now part of Sanas, and presented the deal as a way to move faster into carrier networks and communications platforms.
Sanas said the acquisition would let it deliver its speech technology directly inside the live communication environments that enterprises rely on, rather than only through software installed at the agent's desk.
Shawn Zhang, Sanas's co-founder and chief technology officer, told Speech Technology that the aim was for speech to behave predictably and intelligently inside the systems people depend on. No timetable was given for integrating the two companies' products.
