Alcatel N.v.

Manfred Immendorfer

Dieter Kopp

Thomas Hormann

Helmut Seichter

Heidi Hackbarth

Gebhard Thierer

A method for recognizing spoken words of a speech includes extracting feature vectors from a speech signal which corresponds to a spoken phrase, and segmenting and classifying the successive extracted feature vectors into syllable oriented word subunits by means of a stored supply of word subunits to form a set of hypotheses. The set of hypotheses is used to generate, by three dimensional time dynamic comparision, a set of word hypotheses by comparing the segmented and classified word subunits with standard pronunciations and pronunciation variants of a plurality of words stored in a reference pattern vocabulary. The generated set of word hypotheses are then subjected to syntactic analysis to determine the spoken phrase.

Method of speaker adaptive speech recognition

The present invention is directed to a circuit arrangement for connecting a voice-controlled additional facility associated with a telephone station to a telephone line. According to the invention a telephone station (32) is connected with a voice-controlled additional facility (2). The additional facility (2) performs voice-controlled telephone dialing and/or voice-controlled operation of a videotex terminal. This is implemented by a connection unit (14) inserted in the telephone line (16). The connection unit (14) cooperates with the exchange to permit calls to be initiated, in the known manner, by means of a keypad at the telephone station (32). Any type of telephone station (32) can be connected to the voice-controlled additional facility (2) without any engineering modifications.

Circuit connecting a voice-controlled device to a telephone station and

In a telephone switching system, a multifrequency code receiver unit is replaced by a voice-controlled dial signal receiver unit connected to the switching network via an adapter unit. The voice-controlled dial signal receiver unit contains a voice recognition unit, a voice output unit, and a control unit. A multifrequency code receiver is connected in parallel with the voice-controlled dial signal receiver unit and coupled to the adapter unit. The subscriber can enter dial information and service-feature-control information through the microphone of the handset of a telephone station and/or through a numeric keypad of the station. Instead of being informed of the respective line condition by tone signals, the subscriber receives this information in the form of speech from the voice output unit.

Growing community of inventors

Ditzingen, Germany

Manfred Immendorfer