# US 11,310,608 B2 · Method for training a listening situation classifier for a hearing aid and hearing system > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,310,608 B2 - Document ID: US11310608B2 - Status: Granted - Kind: B2 - Application number: 17/110,509 - Filed: 2020-12-03 - Published: 2021-06-03 - Granted: 2022-04-19 - Earliest claimed: 2019-12-03 - Assignees: Sivantos Pte. Ltd. - Applicants: Sivantos Pte. Ltd. - Inventors: Sven Schoen; Christoph Kukla; Andreas Bollmann - Web page: https://patentgrep.com/patent/US11310608B2 ## CPC - H04R 25/70: {Adaptation of deaf aid to hearing loss, e.g. initial electronic fitting} - G06F 18/24: Classification techniques - G06F 18/41: {Interactive pattern learning with a human teacher} - H04R 25/507: {implemented by neural network or fuzzy logic} - G06F 2218/12: Classification; Matching - H04R 2225/41: Detection or adaptation of hearing aid parameters or programs to listening situation, e.g. pub, forest - H04R: LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; ELECTRIC HEARING AIDS; PUBLIC ADDRESS SYSTEMS (producing sounds with frequency not determined by supply frequency G10K) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) ## Abstract In a method for the training of a listening situation classifier for a hearing aid, a user is presented with a number of acoustic signals by use of a terminal device, which prompts the user to indicate the signal source of the signal or the particular presented signal. The training data is adapted for the listening situation classifier in dependence on the user's indication of the presented signal or one of the possibly several signals presented and updates the listening situation classifier by use of the training data. ## Claims ### Claim 1 A method for a training of a listening situation classifier for a hearing aid, which comprises the steps of: presenting a user with a plurality of acoustic signals by means of a terminal device; prompting the user to indicate a signal source of a signal or a particular presented signal; adapting training data for the listening situation classifier in dependence on a user's indication of the particular presented signal or one of several signals presented; and updating the listening situation classifier by means of the training data. ### Claim 2 (depends on 1) The method according to claim 1, wherein the particular presented signal or one of the several signals presented forms a characteristic signal for an unknown acoustic situation not known to the listening situation classifier. ### Claim 3 (depends on 2) The method according to claim 2, wherein the characteristic signal for the unknown acoustic situation contains a recording of a real situation. ### Claim 4 (depends on 3) The method according to claim 3, wherein the recording of the real situation is produced when a hearing aid wearer provides a characteristic input for a setting produced in an unsatisfactory manner by the listening situation classifier and/or when the listening situation classifier can only match up the real situation with a trained listening situation by a probability value falling below a limit value. ### Claim 5 (depends on 2) The method according to claim 2, wherein the characteristic signal for the unknown acoustic situation is saved in a central database and is fetched from the central database for its presentation. ### Claim 6 (depends on 2) The method according to claim 2, wherein the unknown acoustic situation in response to an indication of the signal source is added to the training data as a known listening situation. ### Claim 7 (depends on 6) The method according to claim 6, wherein the characteristic signal for the unknown acoustic situation is presented to a number of users and the unknown acoustic situation is only added to the training data as the known listening situation if a predetermined number of the users have agreed on the indication of the characteristic signal as the signal source. ### Claim 8 (depends on 2) The method according to claim 2, wherein the plurality of acoustic signals from known signal sources are presented to the user or a respective user and an indication of the user as to the signal source matched up with a respective signal is used to determine a suitability value, and the characteristic signal for the unknown acoustic situation is only presented to the user if the suitability value determined for the user exceeds a given value. ### Claim 9 (depends on 1) The method according to claim 1, wherein indications of the user or a respective user are supplied to a central database or an additional central database and are evaluated by an algorithm implemented on the central database or the additional central database in regard to a suitability value of the user or the respective user. ### Claim 10 (depends on 3) The method according to claim 3, wherein the recording or a respective recording of the real situation is associated with meta-data containing information about a listening situation matched up by the listening situation classifier, a current position of the hearing aid, a background noise level, an estimated value for a distance from a sound source, a manufacturing date of the hearing aid and/or operation software and/or a number of microphones used for the recording. ### Claim 11 (depends on 10) The method according to claim 10, wherein the recording or a respective recording of the real situation is standardized as the characteristic signal for an unknown listening situation prior to a presentation. ### Claim 12 (depends on 11) The method according to claim 11, wherein during a standardization, the recording or the respective recording is adjusted to a target format in regard to its length and/or its data format. ### Claim 13 (depends on 12) The method according to claim 12, wherein during the standardization a quality value is assigned to the recording or the respective recording on the basis of the meta-data, and wherein only a recording with a sufficiently high quality value is presented to the user or the respective user. ### Claim 14 (depends on 11) The method according to claim 11, wherein multiple recordings are compared in regard to their acoustic content and recordings with sufficiently similar acoustic content are grouped together. ### Claim 15 A hearing system, comprising: at least one terminal device; a listening situation classifier; and a controller configured to perform a method for a training of a listening situation classifier for the hearing system, said controller programmed to: present a user with a plurality of acoustic signals by means of said at least one terminal device; prompt the user to indicate a signal source of a signal or a particular presented signal; adapt training data for the listening situation classifier in dependence on a user's indication of the particular presented signal or one of several signals presented; and update the listening situation classifier by means of the training data. ### Claim 16 (depends on 15) The hearing system according to claim 15, further comprising a hearing aid with a signal processor, on which the listening situation classifier is implemented.