Leveraging blowing as a directly controlled interface

Breathing is a natural and directly controllable human interaction method. Currently, only a few works have considered breath as a direct input controlling mechanism. The equipment relied upon in these works is generally complicated, expensive, inconvenient to wear, and sometimes insufficiently controllable. This paper proposes a simple method to implement a natural and directly controllable interaction based on blowing air. The method can provide interaction operations to applications by transferring blowing of air into sound using headset microphones and classifying them into different categories of blowing air in conjunction with support vector machines and particle swarm optimization. During tests, it's proved that this interface not only increases the type of blowing used for interaction, but also eliminates interference from speaking in a normal volume effectively. This user interface can be conveniently used in special environments (e.g., noise, unavailability of hands) or for special groups (e.g., deaf-mutes).