EP-4742234-A2 - ENVIRONMENTALLY ADAPTIVE MASKING SOUND

Abstract

One aspect of the present disclosure relates to a method comprising: measuring environmental sound proximate to an audio device and outputting a masking sound at the audio device, wherein the masking sound includes a repeating masking sound. The method further comprises adjusting content of the masking sound based on the measured environmental sound, wherein adjusting the content of the masking sound includes selecting the content of the masking sound to mitigate detectability of looping artefacts in the repeating masking sound.

Inventors

• BLEWETT, MARK RAYMOND
• HUANG, CHUAN-CHE
• GAUGER, DANIEL M.
• KEMMERER, JEREMY
• QUAN, Xiao
• BANAR, BERKER

Assignees

• Bose Corporation

Dates

Publication Date

20260513

Application Date

20231002

Claims (15)

1. A method comprising: measuring environmental sound (100) proximate to an audio device (10); outputting a masking sound (110) at the audio device, wherein the masking sound includes a repeating masking sound; and adjusting content of the masking sound based on the measured environmental sound, wherein adjusting the content of the masking sound includes selecting the content of the masking sound (110) to mitigate detectability of looping artefacts in the repeating masking sound.
2. The method of claim 1, further including adjusting at least one of a volume of the masking sound or a spectrum of the masking sound.
3. The method of claim 1 or 2, further including determining the content of the masking sound based on a model (90) of environmental sound, wherein determining the content of the masking sound includes selecting, by the model of environmental sound, an anti-looping effect to mitigate detectable looping artefacts in the output.
4. The method of any one of claims 1 to 3, further including: using a model (90) of environmental sound in identifying salient sound sources in the environmental sound (100); predicting characteristics of the salient sound sources using the model (90); and determining the masking sound (110) to effectively mask a salient sound source of the salient sound sources, wherein the masking sound (110) is determined based on a predicted characteristic of the salient sound source of the salient sound sources that persists over an extended period of at least several minutes.
5. The method of claim 3 or 4, further including training the model using at least one of a database of environmental sound inputs or a database of instrument and synthesizer sounds.
6. The method of any one of claims 3 to 5, wherein the model is periodically updated based on the measured environmental sound at the audio device.
7. The method of any one of claims 3 to 6, further including determining at least one of a volume of a masking sound or a spectrum of the masking sound based on the model of environmental sound.
8. The method of claim 2 or 7, wherein the volume of the masking sound is determined based on a subset of frequencies in the measured environmental sound.
9. The method of any one of claims 1 to 8, further including determining the content of the masking sound using generative music techniques to generate and/or modify the content of the masking sound (110), thereby disguising or smoothing the looping artefacts in the masking sound (110).
10. The method of any one of claims 1 to 9, wherein the content of the masking sound is determined based on a subset of frequencies in the measured environmental sound.
11. The method of any one of claims 1 to 10, further including initiating a masking sound mode in response to a trigger prior to outputting the masking sound, wherein after initiating the masking sound mode, the content of the masking sound is selected based on the measured environmental sound prior to outputting the masking sound.
12. The method of claim 11, wherein the trigger includes at least one of: a location-based trigger, a user profile trigger, user actuation at an interface connected with the audio device, detection of an acoustic signature in the environmental sound, proximity between the audio device and another device, or a scheduled time trigger.
13. The method of any one of claims 1 to 12, wherein the masking sound (110) that includes the repeating masking sound includes one of a repeating sound of ocean waves, white noise variation, or breeze passing through a forest.
14. An audio device (10) comprising: at least one electro-acoustic transducer (20); a set of microphones (70); and a controller (30) coupled with the at least one electro-acoustic transducer and the set of microphones, the controller configured to: measure environmental sound (100) proximate to the audio device; output a masking sound (110) to the at least one electro-acoustic transducer, wherein the masking sound includes a repeating masking sound; and adjust content of the masking sound based on the measured environmental sound, wherein adjusting the content of the masking sound includes selecting the content of the masking sound (110) to mitigate detectability of looping artefacts in the repeating masking sound.
15. The method of claim 14, wherein the controller is further configured to adjust at least one of a volume of the masking sound or a spectrum of the masking sound.

Description

PRIORITY CLAIM This application claims priority to US Patent Application No. 17/959,462 filed on October 4, 2022, which is incorporated by reference in its entirety. TECHNICAL FIELD This disclosure generally relates to audio devices and related approaches. More particularly, the disclosure relates to applying masking sounds in audio devices. BACKGROUND Masking sounds such as white noise, nature sounds, or chants have various beneficial applications. However, devices for applying conventional masking sounds fail to account for factors in the environmental sound around the user and can fail to provide adequate masking effects. SUMMARY All examples and features mentioned below can be combined in any technically possible way. Various implementations include adaptive masking of environmental sound. Particular implementations are configured to adjust a masking sound based on one or more detected environmental sound sources. In some particular aspects, a method includes: measuring environmental sound proximate to an audio device, outputting a masking sound at the audio device, and adjusting at least one of a volume of the masking sound, a spectrum of the masking sound, or content of the masking sound based on the measured environmental sound. In some particular aspects, an audio device includes: at least one electro-acoustic transducer; a set of microphones; and a controller coupled with the at least one electro-acoustic transducer and the set of microphones, the controller configured to: measure environmental sound proximate to the audio device, output a masking sound to the at least one electro-acoustic transducer, and adjust at least one of: a volume of the masking sound, a spectrum of the masking sound, or content of the masking sound based on the measured environmental sound. Implementations may include one of the following features, or any combination thereof. In certain implementations, the method further includes determining at least one of the volume of the masking sound, the spectrum of the masking sound, or the content of the masking sound based on a model of environmental sound. In particular cases, the model of environmental sound includes a machine learning model, a saliency model, and/or a linear blind source separation model. In some aspects, the model of environmental sound includes a single model or multiple models. In particular implementations, the method further includes training the model using at least one of a database of environmental sound inputs or a database of instrument and synthesizer sounds. In some aspects, the model is periodically updated based on the measured environmental sound at the audio device. In certain cases, the volume of the masking sound is determined based on a subset of frequencies in the measured environmental sound. In particular aspects, the content of the masking sound is determined based on a subset of frequencies in the measured environmental sound. In some implementations, the spectrum of the masking sound is determined based on a subset of frequencies in the measured environmental sound. In some cases, the method further includes initiating a masking sound mode in response to a trigger prior to outputting the masking sound, wherein after initiating the masking sound mode, the volume of the masking sound, the spectrum of the masking sound, or the content of the masking sound is selected based on the measured environmental sound prior to outputting the masking sound. In certain implementations, the trigger includes at least one of: a location-based trigger, a user profile trigger, user actuation at an interface connected with the audio device, detection of an acoustic signature in the environmental sound, proximity between the audio device and another device, or a scheduled time trigger. In particular cases, the scheduled time trigger is according to a user-defined schedule, a collaboratively suggested schedule based on a group of users, and/or a schedule developed by an artificial intelligence engine based on prior user trigger(s) and/or use pattern(s). In some cases, the volume of the masking sound, the spectrum of the masking sound, or the content of the masking sound is selected to mask acoustic energy of at least one salient sound in the measured environmental sound. In particular aspects, the volume of the masking sound is selected based on a sound pressure level (SPL) of at least one salient sound source in the measured environmental sound. In certain cases, the volume of the masking sound is selected to approximately match or exceed an upper range of the SPL of the salient sound source(s), or the volume of the masking sound is set to a defined level such as a defined value or percentage variation (e.g., +/- X decibels (dB) or +/- Y percent dB) from an upper range of the SPL of the salient sound source(s). In certain cases, the SPL of the at least one salient sound is predicted using the measured environmental sound as an input to a model of environment