| 摘要 |
PURPOSE: To learn a standard voice by collating and comparing standard signals stored in a storage means for previously storing the standard signals based upon the frequency bands of 1st formant and 2nd formants of standard voice data with feature signals obtained from an analytical means, judging the voice and displaying the judged result as a picture. CONSTITUTION: Voice signals vocalized by a trainee are continuously extracted by an extraction means 3, the frequency bands of the 1st and 2nd formats of the voice are detected by the analytical means 7 and feature signals are continuously obtained. On the other hand, standard signals based upon the frequency bands of the 1st and 2nd formats of standard voice data are previously stored in the storage means 5, the feature signals are collated/compared with the standard signals and judgement signals are outputted. A video processor 8 converts all the continuous feature signals into video signals to be simultaneously displayed and converts the standard signal also into a video signal in order to simultaneously display it. Consequently, all the continuous feature signals based upon the trainee's voice are displayed on a display device 9 together with the standard signals. |
