Feasibility Evaluation for Keyword Spotting System Using Mini Microphone Array on UAV

This paper evaluates the use of mini microphone array, Re-Speaker Core V2.0 for keyword spotting system in the Unmanned Aerial Vehicle (UAV) that is used for disaster management task. We do the recording experiment by simulating help call in a various situation and assess the performance of our keyword spotting model. We also do processing of the recored speech to improve the intelligibility of the speech and compare the performance to the raw recorded data. Our evaluation shows that the use of the said mini microphone array is feasible in a up to a certain distance. We also shows that by employing processing to the recorded sound, we could improve the detection rate of the help call for every situation.

[1]  Tsuyoshi Usagawa,et al.  Contextual keyword spotting in lecture video with deep convolutional neural network , 2017, 2017 International Conference on Advanced Computer Science and Information Systems (ICACSIS).

[2]  Hiroshi G. Okuno,et al.  Design of UAV-Embedded Microphone Array System for Sound Source Localization in Outdoor Environments † , 2017, Sensors.

[3]  Katsutoshi Itoyama,et al.  Noise correlation matrix estimation for improving sound source localization by multirotor UAV , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[4]  H. Aglan,et al.  Spectral Subtraction of Noise in Speech Processing Applications , 2008, 2008 40th Southeastern Symposium on System Theory (SSST).

[5]  Jiri Malek,et al.  Single channel speech enhancement using convolutional neural network , 2017, 2017 IEEE International Workshop of Electronics, Control, Measurement, Signals and their Application to Mechatronics (ECMSM).

[6]  Ian F. Akyildiz,et al.  Help from the Sky: Leveraging UAVs for Disaster Management , 2017, IEEE Pervasive Computing.

[7]  Eric Bavu,et al.  A distributed network of compact microphone arrays for drone detection and tracking , 2017 .

[8]  Hiroshi G. Okuno,et al.  Development of microphone-array-embedded UAV for search and rescue task , 2017, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[9]  Jingxuan Sun,et al.  A Camera-Based Target Detection and Positioning UAV System for Search and Rescue (SAR) Purposes , 2016, Sensors.

[10]  Keisuke Nakamura,et al.  Outdoor auditory scene analysis using a moving microphone array embedded in a quadrocopter , 2012, 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems.