The invention discloses an interactive self-fitting method for an intelligent
hearing aid, and the method comprises the following steps: 1, a
hearing impaired patient listening to test voice emitted by the
hearing aid for five-stage evaluation, and forming a
hearing aid sound quality evaluation index; 2, forming a feature sequence according to the 11-dimensional
audiogram, the 3-dimensional personalized information and the voice evaluation of the patient, and then calculating the similarity between the feature sequence and the feature sequence in the fitting historical
database; 3, designing a value function of parameter adjustment, and calculating the value of a parameter adjustment strategy, so that the value of each parameter adjustment is maximized; 4, constructing a neural network, and obtaining the weight of the neural network; and 5, inputting the test
voice frequency spectrum, the voice evaluation, the patient personality information and three continuous groups of historical parameter adjustment strategies into the neural network, and regenerating the test voice to the
hearing impaired patient for evaluation. The subjective and objective voice evaluation method is comprehensively considered, the
deep learning network is combined,
cognition and personal factors are integrated into the hearing aid design, and the application prospect is good.