A Compressive Sensing-based Speech Signal Processing System for Wearable Computing Device in IPTV Environment
Kuei-Hong Lin, Cheng-Hsun Lin, Kuo-Huang Chung, Kai-Shun Lin · Advances in intelligent systems research/Advances in Intelligent Systems Research · 2013
Abstr act A design of compressive sensing (CS)-based speech signal processing (SSP) system for wearable computing device (WCD) in internet protocol television (IPTV) environment is proposed in this study.In WCD, the voice control would be an important function since the WCD is too small to provide typing device.Hence, the quality of the recorded user's comments will determine the user experience (UX) of the interactions between the user and the WCD.For building a user-friendly entertainment environment of the IPTV, the proposed SSP system integrates the WCD to be the medium of the voice control, that is, the voice recorder of the WCD would record the background voice for further analyzing continuously, and it would waste lots of the bandwidth and power of the WCD to transmit the data to the cloud.Therefore, the proposed method is applied to compress the recorded voice for saving the transmitting bandwidth and power of the WCD via the design of an acoustic echo cancellation (AEC)-aided CS-based SSP technique.From the experimental results, the voice data can be compressed and well-recovered efficiently via the proposed CS-based SSP system.