Back to results
US 12,587,798 B2Granted
Headphones with sound-enhancement and integrated self-administered hearing test
AssigneeAlex Lau
InventorsAlex Lau, Pui Tong Paul Lee
Life of this patent
Estimating the term…
Abstract
The present invention provides a self-contained headphone that simplifies the process of self-administered hearing tests in audiological profiling for sound enhancement; a specially-configured computing device that allows interoperability with headphones/speakers and other transducer components of any make and model, yet capable of audiological profiling for sound enhancement; and the machine learning-based classifiers for predicting audiological profiles and audio settings for a listener using her personal traits, the listening conditions, and the listening contents, without the need for first conducting any hearing test.
Claims11
A self-contained headphone with sound enhancement and integrated hearing profiling, comprising:
- one or more processors and one or more user interface elements including one or more of buttons, sliders, and dials integrated with the headphone for facilitating a self-administered hearing test during hearing profiling of a listener and calibration of the self-contained headphones;
- a personal trait-based audiological profile classifier, implementable by one or more neural networks, and configured to:
- machine-learn from training data comprising records of a plurality of listeners' personal trait features collected by the self-contained headphone or other audio devices, the personal trait features including age, gender, profession, education level, lifestyle, health condition, and the listeners' audiological profiles corresponding to the personal traits;
- classify a new listener based on input personal traits;
- and
- recommend an audiological profile without the self-administered hearing test;
- a listening condition-based audiological profile and audio settings classifier, implementable by one or more neural networks, and configured to:
- machine-learn from training data comprising records of a plurality of listening condition features collected by the self-contained headphone or other audio devices, the listening condition features including ambience noise levels, corresponding environment types, corresponding time and location data of when and where the listening conditions are collected, corresponding listeners' audio settings comprising one or more of playback volume, equalizer settings, noise-cancellation level, and corresponding listeners' selections of audiological profile;
- classify a run-time listening condition;
- and
- recommend an audiological profile and audio settings without the self-administered hearing test;
- a listening content-based audiological profile and audio settings classifier, implementable by one or more neural networks, and configured to:
- machine-learn from training data comprising records of a plurality of listening content features collected by the self-contained headphone or other audio devices, the listening content features including playback track titles, listening content types, corresponding listeners' audio settings comprising one or more of playback volume, equalizer settings, noise-cancellation level, and corresponding listeners' selections of audiological profile;
- classify a run-time listening content;
- and
- recommend an audiological profile and audio settings without the self-administered hearing test;
- and
- a prediction result merger configured to:
- merge the recommended audiological profile recommended by the personal trait-based audiological profile classifier, the recommended audiological profile recommended by the listening condition-based audiological profile and audio settings classifier, and the recommended audiological profile recommended by the listening content-based audiological profile and audio settings classifier to generate a merged audiological profile for the sound enhancement;
- and
- merge the recommended audio settings recommended by the listening condition-based audiological profile and audio settings classifier and the recommended audio settings recommended by the listening content-based audiological profile and audio settings classifier to generate merged audio settings for configuring the self-contained headphone's audio settings.
18 more elements
An apparatus for providing sound enhancement by audiological profiling, comprising:
- one or more processors and a user interface for facilitating a self-administered hearing test during hearing profiling of a listener and calibration of an audio signal generation circuitry path;
- a personal trait-based audiological profile classifier, implementable by one or more neural networks, and configured to:
- machine-learn from training data comprising records of a plurality of listeners' personal trait features collected by the self-contained headphone or other audio devices, the personal trait features including age, gender, profession, education level, lifestyle, health condition, and the listeners' audiological profiles corresponding to the personal traits;
- classify a new listener based on input personal traits;
- and
- recommend an audiological profile without the self-administered hearing test;
- a listening condition-based audiological profile and audio settings classifier, implementable by one or more neural networks, and configured to:
- machine-learn from training data comprising records of a plurality of listening condition features collected by the self-contained headphone or other audio devices, the listening condition features including ambience noise levels, corresponding environment types, corresponding time and location data of when and where the listening conditions are collected, corresponding listeners' audio settings comprising one or more of playback volume, equalizer settings, noise-cancellation level, and corresponding listeners' selections of audiological profile;
- classify a run-time listening condition;
- and
- recommend an audiological profile and audio settings without the self-administered hearing test;
- a listening content-based audiological profile and audio settings classifier, implementable by one or more neural networks, and configured to:
- machine-learn from training data comprising records of a plurality of listening content features collected by the self-contained headphone or other audio devices, the listening content features including playback track titles, listening content types, corresponding listeners' audio settings comprising one or more of playback volume, equalizer settings, noise-cancellation level, and corresponding listeners' selections of audiological profile;
- classify a run-time listening content;
- and
- recommend an audiological profile and audio settings without the self-administered hearing test;
- and
- a prediction result merger configured to:
- merge the recommended audiological profile recommended by the personal trait-based audiological profile classifier, the recommended audiological profile recommended by the listening condition-based audiological profile and audio settings classifier, and the recommended audiological profile recommended by the listening content-based audiological profile and audio settings classifier to generate a merged audiological profile for the sound enhancement;
- and
- merge the recommended audio settings recommended by the listening condition-based audiological profile and audio settings classifier and the recommended audio settings recommended by the listening content-based audiological profile and audio settings classifier to generate merged audio settings for configuring the audio signal generation circuitry path's audio settings;
- wherein the apparatus is configured to couple with a headphone or speaker of any make and model and provide the sound enhancement on an input audio signal according to an audiological profile to generate a modified audio signal specific to the audio signal generation circuitry path that includes the headphone or speaker.
19 more elements
Dependencies are inferred from the claim text.