The invention relates to a method for determining at least one relevant single image, wherein a plurality of single optical images are generated during a continuous optical measurement (3) of a dental subject (1) to be recorded. During the optical measurement (3) an audio recording is generated by means of a microphone (4) and at least one speech sequence (12； 15； 28) spoken by a user (6) is recorded. The relevant single image (7； 8； 24) is then selected within a specified sequence duration between a start time (17, 18) and an end time of the speech sequence (12； 15； 28).
