At present,the detection of stuttering in China is mainly through the subjective evaluation of language experts,and there is a lack of intelligent and objective medical aids to detect the severity of stuttering.In response to this phenomenon,this pa-per evaluates the severity of children's stuttering based on the UClass stuttering corpus,and uses the Faster-Rcnn-Fpn deep learn-ing algorithm to detect the spectrogram.The experimental results show that the model can effectively detect stuttering speech repeti-tion,prolongation and interjection types,and calculate the language efficiency score(SES)according to its duration to express the severity of stuttering.It is expected to become a medical auxiliary tool for calculating and analyzing the severity of stuttering,which is convenient for early detection of children's stuttering disorder and helps children's physical and mental development.
关键词
深度学习/Faster-Rcnn-Fpn算法/口吃/目标检测/语谱图识别
Key words
deep learning/Faster-Rcnn-Fpn algorithm/stuttering/object detection/spectrogram recognition
维普
万方数据