Automatic Speech Recognition (ASR) is a valid solution for smartphone and PC interaction from people with motor skill impairments. However recognition performances are reduced in scenarios affected by environmental noise. We propose a speech enhancement system based on MEMS microphone array and a digital signal processor capable to increase signal-to-noise ratio (SNR) of the user’s voice. The difference in Time of Arrivals (TOA) audio samples between microphones is exploited by the array to separate speech and environmental noise. In such way the system can obtain an increment in SNR about 16.5 dB using an Adaptive Noise Reduction filter. Due to the small array size, the proposed system has been integrated in a wearable device. Theoretical analysis and in-system measurements prove the effectiveness of the proposed solution.
Wearable speech enhancement system for motor impaired people
PALLA, ALESSANDRO;FANUCCI, LUCA;
2017-01-01
Abstract
Automatic Speech Recognition (ASR) is a valid solution for smartphone and PC interaction from people with motor skill impairments. However recognition performances are reduced in scenarios affected by environmental noise. We propose a speech enhancement system based on MEMS microphone array and a digital signal processor capable to increase signal-to-noise ratio (SNR) of the user’s voice. The difference in Time of Arrivals (TOA) audio samples between microphones is exploited by the array to separate speech and environmental noise. In such way the system can obtain an increment in SNR about 16.5 dB using an Adaptive Noise Reduction filter. Due to the small array size, the proposed system has been integrated in a wearable device. Theoretical analysis and in-system measurements prove the effectiveness of the proposed solution.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
