Speaker identification for personalized smart TVs
Wei-Ho Tsai, Jui-Che Lin, Cin-Hao Ma, Yuan‐Fu Liao · 2016
Personalized service is an important feature for smart TVs. To facilitate personalized services, identifying the user is a necessity. This study investigates how to identify a smart TV user from his/her voice, that is, speaker identification (SID). Recognizing the fact that when a user issues a voice command to a smart TV, the signal received by the smart TV would not be only the user's speech but also the background sound mainly from the TV. Sometimes the background sound can be louder than the user's speech, and hence it is detrimental for SID. To improve the SID, we propose two compensating approaches built upon the spectrum subtraction. Our experiment show that both of the proposed compensating approaches improves the SID accuracy noticeably.