A Set of DSP System to Detect Baby Crying

摘要

In order to improve the quality of infant health care, the function of baby crying recognition is proposed to be added to the nursery box. The hardware core is consist of digital signal processor (DSP) chip TMS320DM643 and multimedia audio codec chip TLC320AIC23B, which will perform well in real time to recognize baby crying. After the baby crying signal is collected by pickups, it will be processed by the audio decoder chip and then sent it to the DSP chip. In the DSP chip, the sound signal will be preprocessed and extracted for the characteristic parameters - linear prediction coefficient (LPC) by optimized autocorrelation function algorithm. It will be recognized accurately by the method of dynamic time regular (DTW) recognition algorithm, and then the results are sent to the host computer through the serial port. It is proved that the accuracy of the baby crying state can be recognized as high as 97.1%, this study is of great importance in the field of infant care.