# US 2023/0054864 A1 · Methods and Apparatus for Audio Equalization Based on Variant Selection

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0054864 A1
- Document ID: US20230054864A1
- Status: Published application
- Kind: A1
- Application number: 18/048,615
- Filed: 2022-10-21
- Published: 2023-02-23
- Earliest claimed: 2019-11-26
- Assignees: Gracenote, Inc.
- Applicants: Gracenote, Inc.
- Inventors: Robert Coover; Joseph Renner; Cameron A. Summers
- Web page: https://patentgrep.com/patent/US20230054864A1

## CPC

- G06N 3/08: Learning methods
- G10H 1/12: by filtering complex waveforms (G10H1/14, G10H1/16 take precedence)
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/0499: Feedforward networks
- G06N 3/09: Supervised learning
- H03G 5/005: {of digital signals}
- H03G 5/165: {Equalizers; Volume or gain control in limited frequency bands}
- G06F 3/165: {Management of the audio stream, e.g. setting of volume, audio stream path}
- G10H 2210/036: of musical genre, i.e. analysing the style of musical pieces, usually for selection, filtering or classification
- G10H 2250/311: Neural networks for electrophonic musical instruments or musical processing, e.g. for musical recognition or control, automatic composition or improvisation
- G10L 25/30: using neural networks
- G10L 25/51: for comparison or discrimination
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G10H: ELECTROPHONIC MUSICAL INSTRUMENTS; INSTRUMENTS IN WHICH THE TONES ARE GENERATED BY ELECTROMECHANICAL MEANS OR ELECTRONIC GENERATORS, OR IN WHICH THE TONES ARE SYNTHESISED FROM A DATA STORE
- H03G: CONTROL OF AMPLIFICATION
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

Methods, apparatus, systems and articles of manufacture are disclosed methods and apparatus for audio equalization based on variant selection. An example apparatus includes a processor to obtain training data, the training data including a plurality of reference audio signals each associated with a variant of music and organize the training data into a plurality of entries based on the plurality of reference audio signals, a training model executor to execute a neural network model using the training data, and a model trainer to train the neural network model by updating at least one weight corresponding to one of the entries in the training data when the neural network model does not satisfy a training threshold.

## Claims

### Claim 1

A computing device comprising:
one or more processors; and
a non-transitory, computer-readable medium storing instructions that, when executed by the one or more processors, cause the computing device to perform a set of operations comprising:
determining that a neural network model does not satisfy a training threshold;
training a neural network model by updating at least one weight corresponding to at least one entry in training data, wherein the training data includes a plurality of reference audio signals each associated with a variant of music; and
transmitting the trained neural network model to a media unit.

### Claim 2 (depends on 1)

The computing device of claim 1, wherein determining that the neural network does not satisfy a training threshold comprises determining that a predetermined accuracy threshold of the training data has not been satisfied.

### Claim 3 (depends on 1)

The computing device of claim 1, wherein the variant of music is a music genre.

### Claim 4 (depends on 1)

The computing device of claim 1, wherein the variant of music is a music volume.

### Claim 5 (depends on 1)

The computing device of claim 1, wherein the variant of music is one or more of: (i) a music mood; (ii) a music tempo; and (iii) a music tempo.

### Claim 6 (depends on 1)

The computing device of claim 1, wherein the media unit is media unit in an automobile.

### Claim 7 (depends on 1)

The computing device of claim 1, wherein transmitting the trained neural network model to a media unit comprises transmitting the trained neural network model to a media unit when the neural network model satisfies the training threshold.

### Claim 8 (depends on 1)

The computing device of claim 1, wherein the set of operations further comprises:
prior to determining that a neural network model does not satisfy a training threshold, obtaining the training data; and
organizing the training data into a plurality of entries based on the plurality of reference audio signals.

### Claim 9

A non-transitory computer readable storage medium comprising instructions which, when executed, cause at least one processor to at least:
determine that a neural network model does not satisfy a training threshold;
train a neural network model by updating at least one weight corresponding to at least one entry in training data, wherein the training data includes a plurality of reference audio signals each associated with a variant of music; and
transmit the trained neural network model to a media unit.

### Claim 10 (depends on 9)

The non-transitory computer readable storage medium of claim 9, wherein the variant of music is a music genre.

### Claim 11 (depends on 9)

The non-transitory computer readable storage medium of claim 9, wherein the variant of music is a music volume.

### Claim 12 (depends on 9)

The non-transitory computer readable storage medium of claim 9, wherein the variant of music is one or more of: (i) a music mood; (ii) a music tempo; and (iii) a music tempo.

### Claim 13 (depends on 9)

The non-transitory computer readable storage medium of claim 9, wherein the media unit is media unit in an automobile.

### Claim 14 (depends on 9)

The non-transitory computer readable storage medium of claim 9, further comprising instructions which, when executed, cause at least one processor to at least:
prior to determining that a neural network model does not satisfy a training threshold, obtain the training data; and
organize the training data into a plurality of entries based on the plurality of reference audio signals.

### Claim 15

A method comprising:
determining that a neural network model does not satisfy a training threshold;
training a neural network model by updating at least one weight corresponding to at least one entry in training data, wherein the training data includes a plurality of reference audio signals each associated with a variant of music; and
transmitting the trained neural network model to a media unit.

### Claim 16 (depends on 15)

The method of claim 15, wherein the variant of music is a music genre.

### Claim 17 (depends on 15)

The method of claim 15, wherein the variant of music is a music volume.

### Claim 18 (depends on 15)

The method of claim 15, wherein the variant of music is one or more of:
(i) a music mood; (ii) a music tempo; and (iii) a music tempo.

### Claim 19 (depends on 15)

The method of claim 15, wherein the media unit is media unit in an automobile.

### Claim 20 (depends on 15)

The method of claim 15, further comprising:
prior to determining that a neural network model does not satisfy a training threshold, obtaining the training data; and
organizing the training data into a plurality of entries based on the plurality of reference audio signals.