Imagine placing your smartphone on a table in a noisy restaurant and clearly capturing the voices of friends seated around you, or recording a lecturer’s voice with clarity in a reverberant auditorium. We introduce SonicSieve, the first intelligent directional speech extraction system for smartphones using a bio-inspired acoustic microstructure. Our passive design embeds directional cues onto incoming speech without any additional electronics. It attaches to the in-line mic of low-cost wired earphones which can be attached to smartphones. We present an end-to-end neural network that processes the raw audio mixtures in real-time on mobile devices. Our results show that SonicSieve achieves a signal quality improvement of 5.0 dB when focusing on a 30° angular region. Additionally, the performance of our system based on only two microphones exceeds that of conventional 5-microphone arrays.

Creating personalized 3D printing objects has become increasingly popular with the advance in end-user modeling tools and fabrication techniques. However, it remains challenging for novice makers to design and fabricate functional objects with mechanical motion such as linear or rotational. To empower users to add mechanical movements to their models for particular needs, we first investigate the commonly used 3D printable mechanisms and parameterize four mechanical primitives. We then provide a computational approach to reconfigure the mechanical primitives to user-provided 3D models, with an underlying structural optimization procedure according to a specified bending force. We further showcase a set of design examples and conduct a user study to demonstrate the potential of our approach in creating personalized functional artifacts.

Voice user interfaces (VUI) are becoming indispensable in car for offering drivers the opportunity to make distraction-free inputs and conduct complex tasks. However, the usability and control efficiency of today’s VUI remain to be enhanced due to its sequential nature. In this work, we explored gestural input on the steering wheel to improve the interaction efficiency of VUI. Based on limitations in VUI, we designed novel gestural commands on the steering wheel to augment them. We also elicited corresponding user-defined gestures by exploring drivers’ touch behavior. Then, we implemented a prototype attached to the steering wheel for recognizing gestures. Finally, we evaluated our system’s usability regarding driving performance, interaction efficiency, cognitive workload and user feedback. Results revealed that our system improved the control efficiency of VUI and reduced workload without a significant reduction in driving distraction than just using VUI.