User adaptive hearing aid self-fitting method based on neural network

Abstract

A user adaptive hearing aid self-fitting method based on a neural network comprises the following steps: fitting a group of initial signal processing parameters according to an audiogram obtained by pure tone audiometry of a user; performing multiple rounds of speech audiometry and parameter adjustment through a hearing aid which is worn by a user and is pre-adjusted with the signal processing parameters; wherein after each time of speech audiometry is completed, the speech audiometry recognition probability model based on the neural network predicts the recognition probability of the user on different speech audiometry corpora under various parameter groups according to the speech audiometry record of the user; the parameter group optimization system obtains several parameter groups enabling the speech audiometry recognition rate of the user to be the highest for the user to select according to the prediction result, and parameters continue to be finely adjusted based on the parameter groups selected by the user; and performing speech audiometry and parameter adjustment continuously and iteratively until the hearing aid is deployed to meet a predetermined hearing aid effect. According to the invention, the fitting of the hearing aid is convenient, and the accuracy of the fitting result is improved at the same time.

Description

technical field [0001] The invention relates to a fitting technology of a digital hearing aid, in particular to a neural network-based user-adaptive hearing aid self-fitting method. Background technique [0002] The fitting of traditional digital hearing aids needs to go to a professional fitting center with the help of experts and professional equipment. Such a matching process is cumbersome and expensive. In recent years, with the development of related technologies, more and more user-independent fitting methods without the intervention of experts and professional equipment have been proposed. The current self-fitting methods mainly have the following problems: (1) The preset parameter set cannot be accurately Match user needs. (2) The convenience of the fitting process and the accuracy of the fitting results cannot be guaranteed at the same time. (3) Rely on the subjective feedback of the user, and do not make full use of the objective listening results of the user. ...

AI Technical Summary

Problems solved by technology

Such a fitting process is cumbersome and costly

In recent years, with the development of related technologies, more and more user-independent fitting methods without the intervention of experts and professional equipment have been proposed. The current self-fitting methods mainly have the following problems: (1) The preset parameter set cannot be accurate Match user needs

(2) The convenience of the fitting process and the accuracy of the fitting results cannot be guaranteed at the same time

(3) Relying on the subjective feedback of the user, not making full use of the objective listening results of the user

[0008] 1) The method based on the preset parameter group cannot guarantee that the preset options meet the needs of users

[0009] 2) The existing self-fitting method based on user fine-tuning does not make full use of the audiometry feedback information of the user during the fitting process, and cannot obtain a good balance between the accuracy of the fitting result and the convenience of the fitting process

[0010] 3) The current self-fitting method based on deep learning is too dependent on user subjective feedback, which does not have objectivity and reliability

It also doesn't take full advantage of the user's objective listening results

Images