摘要 |
PURPOSE:To offer the practical voice recognizing method being excellent in the noise resistance, which is operated even under such a noise environment as the peripheral noise environment is varied together with time. CONSTITUTION:According to the voice recognizing method by a neural network model, a noise elimination processing 17 is performed to a voice to which various noises are superposed, and by allowing the neural network model 5 to learn a voice pattern which contains the noise which cannot be eliminated, an essential feature of the voice is learned well, and the voice recognition of high accuracy which is scarcely influenced by a peripheral noise. |