Bernhard Kaemmerer

Computer voice recognition method verifying speaker identity using speaker and non-speaker data

A method for recognizing a speaker in which a voice signal is spoken into a computer by a speaker and a feature vector is formed for the voice signal. The feature vector is compared to at least one stored reference feature vector and to at least one anti-feature vector. The reference feature vector is formed from a speech sample of a speaker to be verified. The anti-feature vector was formed from a speech sample that was spoken in by another speaker who is not the speaker to be verified. A 2-class classification is resolved by forming a similarity value and evaluating the similarity value on the basis of a predetermined range within which the similarity value must deviate from a predetermined value so that the voice signal can be classified as deriving from the speaker to be verified.

Method and apparatus for the recognition of individually spoken words

A system for recognizing individually spoken words includes an acousto-electric transducer with a controllable amplifier and a high-pass filter connected to a level calculation path for calculating the level of the output signal by detecting the signal envelope and converting the envelope signal to a 1-bit digital signal to be supplied to a microprocessor. A signal analysis path is also connected to the transducer and includes a low-pass filter followed by a 1-bit analog to digital converter and a digital processing unit which supplies data units to the microprocessor. Individually spoken words are recognized and actions taken accordingly.