Disclosed herein, among other things, are apparatus and methods for automated assessment and adjustment of tinnitus-masker impact on speech intelligibility during fitting. In various embodiments, a method of fitting a tinnitus-masker device for a patient is provided. The method includes using a speech intelligibility model to predict impact of therapy provided by the tinnitus-maker device on speech understanding of the patient. A parameter of the tinnitus-masker device is automatically adjusted during fitting to reduce the impact of the therapy provided by the tinnitus-maker device on speech understanding of the patient, according to various embodiments.
