London-based startup Relay announced in August 2026 the debut of its macOS voice-to-text software, along with plans to release a dedicated portable microphone called the Relay Q in early 2027. The company is positioning dedicated hardware as a solution to a practical problem that existing software-only dictation tools have not fully addressed.
Relay was founded by two former Nothing employees: Cookie Xu, who served as Nothing’s head of AI and services, and Raymond Zhu, a hardware product manager. The pair say their idea grew directly from their own experience using Wispr Flow’s transcription software while working at Nothing. Though they found voice-to-text useful for thinking through ideas, they struggled with a consistent obstacle: finding an appropriate speaking volume in a shared office environment.
“In the office, we never really could get the right volume for input,” Zhu told WIRED. “If I whisper too quietly, the Mac wouldn’t be able to pick it up; too loud and my coworker next to me can hear.” Zhu said he would sometimes hold an earbud in front of his mouth as a workaround, adding, “It calls for dedicated hardware.”
The Relay macOS app is available now in beta and works with or without the company’s forthcoming microphone. Users select a hotkey — the default is Fn — which activates Relay when held while a cursor is in any text field. The voice-to-text engine runs on Google’s Gemini models. Relay says it plans to expand to Windows, iOS, and Android, but is launching on desktop first, citing the greater complexity of work-related communication compared to casual mobile messaging.
The Relay Q hardware remains months away from release. The app’s launch places Relay alongside existing voice-to-text tools such as Wispr Flow and Google’s on-device dictation features, though the company is differentiating itself through its hardware-first approach to audio input.
Source: WIRED