Abstract: Voice command operated systems are being installed in modern motor vehicles with increasing frequency. Such systems should be operable by various vehicle occupants and from various seating positions. A new type of speech recognition system is described, which is good at accomplishing these tasks. Therein, for regulating the speech recognition system, the reception characteristic of a a directionally selective microphone array (12) is controlled by an optimization unit (10). These speech signals are then processed in a speech recognizer (11) at least parallel in time. Then, on the basis of the results provided by the speech recognizer (11), the reception characteristic of the microphone array (12) is so controlled by an optimization unit (10), that the recognition performance of the speech recognizer (11) downstream of the optimization unit (10) is optimized.

Abstract: Voice command operated systems are being installed in modern motor vehicles with increasing frequency. Such systems should be operable by various vehicle occupants and from various seating positions. A new type of speech recognition system is described, which is good at accomplishing these tasks. Therein, for regulating the speech recognition system, the reception characteristic of a a directionally selective microphone array (12) is controlled by an optimization unit (10). These speech signals are then processed in a speech recognizer (11) at least parallel in time. Then, on the basis of the results provided by the speech recognizer (11), the reception characteristic of the microphone array (12) is so controlled by an optimization unit (10), that the recognition performance of the speech recognizer (11) downstream of the optimization unit (10) is optimized.

Abstract: In many real applications such as voice control in vehicles there is the problem that the users change relatively frequently. Then the question arises: which is the correct data set for the current user? The invention provides a process making it possible automatically for the duration of operation of the system to recognize whether the speaker changes, or which (speaker dependent) data set is correct for the actual user. This task is solved by a speech recognition system which is based on a so-called Semi-Continuous Hidden Markov Model (SCHMM). Codebooks are produced, normal distribution is represented, speaker-specific data sets are stored in addition to a so-called base-line data set, and the inventive speech recognition system correlates the speech signal by means of vector quantitization with the speaker-independent and the speaker-dependent codebooks, making it possible to ascertain the identity of the speaker.

Abstract: In a speech recognition method for word sequences that are combined in sentences with defined syntax, syntax information is integrated into a speech model so as to unambiguously identify recurring words through indexing. As a result, the recognition process can be carried out in a manner similar to the situation without integration of the syntax information.