Breath Mobile: A Software-Based Hands-Free and Voice-Free Breathing Controlled Mobile Phone Interface
Jackson Feijó Filho, Thiago Valle, Wilson Prata · 2012 · Proceedings of the 14th International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS) · doi:10.1145/2384916.2384961
Summary
This short paper proposes Breath Mobile, a software-only breathing interface for mobile phones that uses the device's built-in microphone to detect puffing patterns and translate them into navigation and selection events. Unlike traditional sip-and-puff controllers that require dedicated external hardware, Breath Mobile processes audio input from the standard phone microphone to detect three distinct breath events: a single short puff (mapped to a right/forward arrow key), a double short puff (mapped to a left/back arrow key), and a long puff (mapped to an action/select key). The proof of concept was implemented as a native application on a Symbian OS smartphone (S60 platform), leveraging the OS's multitasking, keyboard buffer injection, and microphone audio processing capabilities. The system enables users with motor disabilities to perform basic phone operations such as navigating contacts and making calls entirely through breathing, without using hands or voice. The work is motivated by the limitations of existing alternative interaction methods: speech recognition affects user privacy by broadcasting commands audibly, eye trackers and traditional sip-and-puff devices require expensive additional hardware, and most physiological signal-based interfaces (cardiovascular, electrodermal, brain activity) are designed for desktop computers rather than mobile devices.
Key findings
The proof of concept demonstrated that a software-only approach to breath-based interaction is technically feasible using standard mobile phone hardware. By processing microphone audio levels, the system could distinguish between single short puffs, double short puffs, and long puffs with sufficient reliability to navigate phone menus and initiate calls. The key advantage over existing solutions is the zero additional hardware cost — the system works with the phone's built-in microphone, making it potentially accessible to users who cannot afford specialized assistive devices. The approach also preserves user privacy (unlike speech recognition) and operates silently. The three-event input vocabulary (forward, back, select) provides a minimal but functional navigation scheme suitable for linear menu structures common on mobile phones of that era.
Relevance
This research demonstrates an important principle in assistive technology: leveraging existing device sensors for alternative interaction rather than requiring specialized hardware. While the Symbian platform is now obsolete, the core concept of software-based breath detection through a phone microphone remains relevant and could be implemented on modern smartphones with more sophisticated audio processing capabilities. For accessibility practitioners, the work highlights the accessibility potential of the sensors already present in everyday devices — microphones, accelerometers, cameras — that could serve as low-cost alternative input channels. The emphasis on both hands-free and voice-free interaction addresses an important gap: many motor-impaired users who cannot use touch screens also cannot use speech recognition due to dysarthria or environmental privacy concerns, leaving breath control as one of the few remaining input modalities available.
Tags: motor impairment · breath control · sip-and-puff · mobile accessibility · alternative input · hands-free interaction · voice-free interaction