Genetic algorithm based auditory training
Embodiments of the present invention are generally directed to the use of a genetic algorithm for the purpose of providing progressive and adaptive auditory training (rehabilitation) to a recipient of a hearing prosthesis. In general, the genetic algorithm is used to adapt the training process to automatically increase the difficulty of the training based on recipient feedback and performance. That is, the genetic algorithm progressively removes perceivable sounds from the training process so as to generate groups of sounds that are difficult for a recipient to perceive.
1. A method for auditory training executed by an auditory training system, comprising:
selecting an initial set of auditory training sound tokens from an auditory training library;
presenting the initial set of auditory training sound tokens to a recipient via a hearing prosthesis;
receiving recipient feedback in response to presented sound tokens from the initial set of auditory training sound tokens, wherein the feedback indicates the recipient's perception of the presented sound tokens; and
executing a genetic algorithm to automatically increase the difficulty of the auditory training based on the recipient's perception of the presented sound tokens to create a second set of auditory sound tokens for presentation to the recipient.
2. The method of claim 1 , further comprising:
sequentially presenting the initial set of auditory training sound tokens to the recipient via the hearing prosthesis; and
receiving recipient feedback after each sequentially presented sound token.
3. The method of claim 2 , wherein each sound token in the initial set of auditory training sound tokens corresponds to a known sound, and wherein receiving recipient feedback in response to a first presented sound token comprises:
receiving an indication of whether the recipient correctly perceived the known sound corresponding to the first presented sound token.
4. The method of claim 1 , wherein executing a genetic algorithm to automatically increase the difficulty of the auditory training based on the recipient's perception of the sound tokens comprises:
executing the genetic algorithm to progressively remove perceivable sounds from the auditory training so as to generate one or more sets of auditory training sound tokens that are difficult for the recipient to perceive.
5. The method of claim 3 , further comprising:
presenting a visual representation of a plurality of sounds to the recipient in relation to presentation of a selected sound token;
presenting a request for the recipient to select one of the visual representations that corresponds to the selected sound token; and
receiving a selection from the recipient of one of the visual representations.
6. The method of claim 1 , wherein selecting the initial set of auditory training sound tokens from the auditory training library comprises:
selecting recipient-specific sound tokens added to the auditory training library by a user.
7. The method of claim 1 , wherein selecting the initial set of auditory training sound tokens from the auditory training library comprises:
selecting the initial set of auditory training sound tokens at random from the auditory training library.
8. The method of claim 1 , wherein presenting the initial set of auditory training sound tokens to the recipient via the hearing prosthesis comprises:
presenting the initial set of auditory training sound tokens to the recipient via a cochlear implant.
9. The method of claim 1 , wherein executing a genetic algorithm based on the recipient's perception of the presented sound tokens comprises:
executing an interactive augmented genetic algorithm.