Controlling information system or device using speech recognition, has database containing information segments, with further regions activated if corresponding speech signal is not in first region

摘要

The method involves storing useful information in a database (32), and specifying using a speech signal at least one information segment as a first data segment (W1). The segment is announced via a control output or a control signal for the device. The information segments are arranged in the database so that an initially limited first information region is accessible via the speech signal, to select a specified information signal. A further information region of the database is activated if the information segment (W1) corresponding to the speech signal is not contained in the first information region, and this region only contains information segments that correspond to a single semantic category. An independent claim is included for a further method.