Abstract: A medical testing device comprises an analog-to-digital converter for converting to an electrical digital signal a voice-frequency signal originating in a speech by a human speaker. The A/D converter is connected to a speech analyzing circuit which analyzes the digital signal to determine a value for a preselected first parameter characteristic of the speech, the parameter being taken from the group comprising frequency or tone, amplitude or loudness, and speed. A video camera or other component is provided for generating a video signal encoding an image of the human speaker's face during the speech. The camera is connected to an image analyzing circuit which processes the video signal to determine a value for a preselected second parameter characteristic of an emotional expression of the human speaker during the speech. A text encoder is provided for encoding a text corresponding in essential words to the speech.