Can AI Make Two-Way Radio Hands-free?
The recent Brussels Expo was buzzing with activity as Critical Communications World (CCW) 2025 brought together global mission-critical and business-critical communications professionals.
Attendees had a front-row seat to the latest in two-way radio, from TETRA and P25 to hybrid LMR–broadband solutions. The innovation technology focus was on reliable transmission, novel command centers, and integrating video and AI-driven analytics.
Traditional two-way radio technologies continue to be the backbone of public safety and emergency services. Their core mission is to serve reliable, instant, and secure team communications. Two-way radios now embrace hybrid architectures, intelligent analytics, and layered redundancy, yet some features remain relics. Push-to-talk (PTT) is a vivid example of such outdated technology.
In a world where people casually say, “OK, vacuum, clean under the couch” from the quiet comfort of their kitchen, firefighters — in smoke, noise, and bulky gloves — are genuinely thrilled that “the new speaker mic is easy to hold and press.” Especially when every button push might literally be a matter of life and death.
Two-way radios have been integral to mission-critical operations since World War II. With the introduction of VOX (a switch that activates when sound above a certain threshold is detected), it might have seemed that the problem of hands-free operation was solved. But that wasn’t the case, unfortunately.
Indeed, VOX allows radios to automatically transmit when the user speaks, keeping hands and attention free for the task at hand. While VOX offers hands-free convenience, discussions at CCW also addressed its limitations.
VOX can be a tricky feature, and it isn’t just a matter of turning it on to make it work properly. In real-world conditions, VOX systems are notoriously unreliable, frequently triggering on background noise, heavy breathing, a side conversation, or a loudspeaker carrying the other side of the conversation. On the other hand, VOX can fail to activate when speech is soft or distant. VOX is not suitable for noisy environments, as it doesn’t discriminate between voice and extraneous noise. This results in missed messages, false transmissions, and a degradation of user trust.
This highlights the need for POLYN’s practical solution. Powered by the NeuroVoice neuromorphic AI chip, it delivers intelligent, real-time voice activation with ultra-low power consumption and sub-millisecond latency — ideal for portable communication devices. NeuroVoice employs a neuromorphic analog architecture to enable AI-based voice activity detection (VAD), keyword-triggered operation, and speaker recognition. These, in combination with voice extraction, promise efficient voice operation in noisy environments.
All that is possible at less than 200 µW, within the power envelope of standard headset accessories. This transforms any headset into a smart, tactical-grade voice interface, offering always-on voice management: reliable, PTT-free communication, improved intelligibility even on low-end radios, and extended communication range in noisy or high-interference environments.
It’s no surprise that POLYN’s value proposition sparked high interest at the CCW – it resonates with a real-world market need, where users still rely on PTT. Only a few manufacturers have begun considering “Smart VOX” concepts.
POLYN’s products enable the creation of headsets and speaker microphones with intelligent, context-aware voice management that adapts to individual speech patterns and noise environments. This way, users will soon be able to use AI-powered hands-free radios that are aligned with the real-time demands of critical missions.

