| 摘要 |
The invention relates to a voice recognition system which is robust to echoes and other background noise. An additional input signal describing a disturbance is evaluated such that during this additional recognition there is maximum suppression of information contained in the input signal. For this purpose, comparison vectors are formed which are continuously adapted to the instantaneous interference. The voice recognition system receives speech signals superimposed by noise signals. Additionally, a first spectral analysis unit produces first spectral values combined to first spectral vectors derived from disturbed speech signals. Estimates of the noise signals are also produced. A second spectral analysis unit produces second spectral values combined to second spectral vectors from the noise signal estimates. Reference vectors formed by reference spectral values to which reference vectors are assigned indices and which reference vectors respond to the spectrums of reference speech signals are stored in memory. Comparison vectors are produced and assigned to the indices from the reference vectors and the second spectral vectors, a continuous adaptation being effected of at least one of the comparison vectors to the second spectral vectors. Indices are then selected by comparing the comparison vectors with the first spectral vectors.
|
