Methods and systems for automatically equalizing audio output based on room position

Patent metadata

Field	Value
Publication number	US-2021194447-A1
Application number	US-202117191510-A
Country	US
Kind code	A1
Filing date	Mar 3, 2021
Priority date	Oct 4, 2017
Publication date	Jun 24, 2021
Grant date	—

Abstract

The various implementations described herein include methods, devices, and systems for automatic audio equalization. In one aspect, a method is performed at an electronic device that includes speakers, microphones, processors and memory. The electronic device outputs audio user content from the speakers and automatically equalizes subsequent audio output of the device without user input. The automatic equalization includes: (1) obtaining audio content signals, including receiving outputted audio content at each microphone; (2) determining from the audio content signals phase differences between microphones; (3) obtaining a feature vector based on the phase differences; (4) obtaining a frequency correction from a correction database based on the obtained feature vector; and (5) applying the obtained frequency correction to the subsequent audio output.

First claim

What is claimed is:

1 . A method for equalizing audio at an electronic device located in a defined space, comprising:
outputting first audio content from at least one speaker of the electronic device; capturing, via at least one microphone, audio data representing a set of reflections of the first audio content within the defined space; obtaining, by a processor of the electronic device, a feature vector based on the audio data that was captured; obtaining, by the processor, a frequency correction based on the feature vector; and applying, by the processor, the frequency correction to subsequent audio content.

2 . The method of claim 1 , further comprising:
prior to the outputting:
outputting, via a speaker device positioned at a particular position, training audio,
receiving the outputted training audio,
generating a reference frequency correction based on the outputted training audio, and
storing the reference frequency correction.

3 . The method of claim 1 , wherein the at least one microphone comprises a plurality of microphones, and wherein the method further comprises:
determining a plurality of phase differences of the audio data that was captured between different microphones of the plurality of microphones.

4 . The method of claim 3 , wherein obtaining the feature vector comprises:
obtaining, by the processor, the feature vector based on the plurality of phase differences.

5 . The method of claim 4 , further comprising:
assigning a plurality of weights to the plurality of phase differences such that each phase difference of the plurality of phase differences is assigned a corresponding weight;
and wherein the feature vector is based on the plurality of phase differences with the plurality of weights assigned thereto.

6 . The method of claim 5 , wherein the plurality of weights is based on a signal-to-noise ratio for the audio data that was captured at each microphone of the different microphones of the plurality of microphones.

7 . The method of claim 3 , wherein the plurality of weights is based on relative positioning of the plurality of microphones.

8 . The method of claim 4 , wherein obtaining the feature vector based on the plurality of phase differences comprises:
applying a fast Fourier transform (FFT) to the plurality of phase differences.

9 . The method of claim 1 , wherein the first audio content comprises at least one of: audio content that is selected by a user of the electronic device, a test signal, or training audio.

10 . The method of claim 1 , wherein applying the frequency correction to the subsequent audio content comprises:
adjusting a gain for a particular range of frequencies.

11 . An electronic device located in a defined space, comprising:
at least one speaker; at least one microphone; a memory storing at least one computer program; and a processor interfaced with the at least one speaker, the at least one microphone, and the memory, and configured to execute the at least one computer program to cause the processor to:
cause the at least one speaker to output first audio content,
capture, via the at least one microphone, audio data representing a set of reflections of the first audio content within the defined space,
obtain a feature vector based on the audio data that was captured,
obtain a frequency correction based on the feature vector, and
apply the frequency correction to subsequent audio content.

12 . The electronic device of claim 11 , wherein the at least one microphone comprises a plurality of microphones, and wherein the processor is further configured to:
determine a plurality of phase differences of the audio data that was captured between different microphones of the plurality of microphones.

13 . The electronic device of claim 11 , wherein the processor is further configured to:
prior to obtaining the feature vector, determine that the first audio content comprises audio having a frequency below a transition frequency for the defined space.

14 . The electronic device of claim 11 , wherein the processor is further configured to:
prior to obtaining the feature vector, determine that the first audio content comprises an acoustic energy that meets at least one energy criteria for a particular range of frequencies.

15 . The electronic device of claim 11 , wherein the processor is further configured to:
prior to obtaining the feature vector, determine that the first audio content comprises an audio coherence that meets at least one signal-to-noise criteria.

16 . A non-transitory computer-readable storage medium storing one or more programs, the one or more programs comprising instructions, which when executed by an electronic device located in a defined space and comprising at least one speaker, at least one microphone, a processor, and memory, cause the electronic device to perform operations comprising:
outputting first audio content from the at least one speaker; capturing, via the at least one microphone, audio data representing a set of reflections of the first audio content within the defined space; obtaining a feature vector based on the audio data that was captured; obtaining a frequency correction based on the feature vector; and applying the frequency correction to subsequent audio content.

17 . The non-transitory computer-readable storage medium of claim 16 , wherein the at least one microphone comprises a plurality of microphones, and wherein the operations further comprise:
determining a plurality of phase differences of the audio data that was captured between different microphones of the plurality of microphones.

18 . The non-transitory computer-readable storage medium of claim 16 , wherein the first audio content comprises at least one of: audio content that is selected by a user of the electronic device, a test signal, or training audio.

19 . The non-transitory computer-readable storage medium of claim 16 , wherein applying the frequency correction to the subsequent audio content comprises:
adjusting a gain for a particular range of frequencies.

20 . The non-transitory computer-readable storage medium of claim 16 , wherein the operations further comprise:
prior to obtaining the feature vector, determining that the first audio content comprises audio having a frequency below a transition frequency for the defined space.

Assignees

• Google Llc

Inventors

• Shaya Benjamin Louis
• Larsen Rasmus
• Lyon Richard F
• Smedegaard Michael

Classifications

• G06N3/0499

  Feedforward networks · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/092

  Reinforcement learning · CPC title

• H03G5/165 Primary

  Equalizers; Volume or gain control in limited frequency bands · CPC title

• G06N3/04

  Architecture, e.g. interconnection topology · CPC title