Swiss hearing-care company Phonak has unveiled Roger On Sphere, which it describes as the world's first remote microphone powered by a deep neural network that separates speech from background noise in real time.
The device was announced on October 14, 2026, from Stafa, Switzerland. It is aimed at people with hearing loss who struggle to follow conversations in noisy settings such as restaurants, meetings and family gatherings, where hearing aids alone may not be enough.
How it works
Roger On Sphere combines Phonak's Sphere speech-separation technology with its established Roger wireless protocol. The microphone continuously analyses the listening environment and isolates speech from background noise in every microphone mode and at every noise level, according to the company.
Compared with its predecessor, the new device delivers an additional 3 decibels of signal-to-noise ratio benefit, which Phonak says is equivalent to roughly 50 per cent less noise power, letting users communicate in environments with twice the noise level.
The microphone automatically adapts to different situations, including meetings, lectures, online calls, restaurants, family gatherings and car journeys, so users do not have to adjust settings manually. Key features include real-time AI speech-from-noise separation through the company's RogerSense OS software and low-latency wireless transmission.
"Roger has transformed communication in demanding listening environments for nearly two decades," said Alain Kertesz, Senior Director Roger at Phonak. "Roger On Sphere brings AI-driven speech separation to a remote microphone, giving users greater clarity and confidence in situations that were previously difficult to manage."
The launch marks another step in the hearing-aid industry's push to embed artificial intelligence directly into assistive devices, following the broader trend of on-device machine learning moving beyond smartphones into specialised health technology.
Sources:
Phonak launches Roger On Sphere: the first AI-powered remote microphone to separate speech from noise in real time — The Manila Times (GlobeNewswire)
The device was announced on October 14, 2026, from Stafa, Switzerland. It is aimed at people with hearing loss who struggle to follow conversations in noisy settings such as restaurants, meetings and family gatherings, where hearing aids alone may not be enough.
How it works
Roger On Sphere combines Phonak's Sphere speech-separation technology with its established Roger wireless protocol. The microphone continuously analyses the listening environment and isolates speech from background noise in every microphone mode and at every noise level, according to the company.
Compared with its predecessor, the new device delivers an additional 3 decibels of signal-to-noise ratio benefit, which Phonak says is equivalent to roughly 50 per cent less noise power, letting users communicate in environments with twice the noise level.
The microphone automatically adapts to different situations, including meetings, lectures, online calls, restaurants, family gatherings and car journeys, so users do not have to adjust settings manually. Key features include real-time AI speech-from-noise separation through the company's RogerSense OS software and low-latency wireless transmission.
"Roger has transformed communication in demanding listening environments for nearly two decades," said Alain Kertesz, Senior Director Roger at Phonak. "Roger On Sphere brings AI-driven speech separation to a remote microphone, giving users greater clarity and confidence in situations that were previously difficult to manage."
The launch marks another step in the hearing-aid industry's push to embed artificial intelligence directly into assistive devices, following the broader trend of on-device machine learning moving beyond smartphones into specialised health technology.
Sources:
Phonak launches Roger On Sphere: the first AI-powered remote microphone to separate speech from noise in real time — The Manila Times (GlobeNewswire)
