# US 2020/0288228 A1 · System and earphone related to transformation of sensed data

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

- Number: US 2020/0288228 A1
- Document ID: US20200288228A1
- Status: Published application
- Kind: A1
- Application number: 16/402,251
- Filed: 2019-05-03
- Published: 2020-09-10
- Earliest claimed: 2019-03-08
- Assignees: Merry Electronics(Shenzhen) Co., Ltd.
- Applicants: Merry Electronics(Shenzhen) Co., Ltd.
- Inventors: Yu-Fei Tsai; Yen-Sheng Chen
- Web page: https://patentgrep.com/patent/US20200288228A1

## CPC

- H04R 1/1016: {Earpieces of the intra-aural type}
- H04R 1/1091: {Details not provided for in groups H04R1/1008 - H04R1/1083}
- G06N 20/20: Ensemble learning
- 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)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

A system for transforming sensed data is provided. The system includes an inertial sensor and a processor. The inertial sensor generates first sensed data by sensing a motion or behavior of a user. The processor is communicatively connected to the inertial sensor and pre-stores a data transformation model, wherein the data transformation model converts an acceleration vector and an angular velocity vector corresponding to the first sensed data into an acceleration vector and an angular velocity vector corresponding to second sensed data, and outputs the acceleration vector and the angular velocity vector corresponding to the second sensed data to an identification model pre-stored in the processor.

## Claims

### Claim 1

An earphone, comprising:
a speaker playing audio data;
an inertial sensor sensing a motion of a user to generate first sensed data; and
a processor communicatively connected to the inertial sensor and the speaker and pre-storing at least one identification model and a data transformation model, wherein the data transformation model transforms the first sensed data received from the inertial sensor into second sensed data, and the at least one identification model determines whether the motion is correct by using the second sensed data, so as to generate an output result.

### Claim 2 (depends on 1)

The earphone of claim 1, wherein the at least one identification model is updated according to the second sensed data.

### Claim 3 (depends on 1)

The earphone of claim 1, wherein the second sensed data comprises an acceleration vector and an angular velocity vector different from those in the first sensed data.

### Claim 4 (depends on 1)

The earphone of claim 1, wherein the processor further pre-stores a reference acceleration vector, and updates one of the first sensed data and the second sensed data based on a vector angle between an acceleration vector in the first sensed data and the reference acceleration vector.

### Claim 5 (depends on 1)

The earphone of claim 1, wherein the processor further pre-stores a reference acceleration vector, determines that the earphone is worn at an incorrect position based on a vector angle between an acceleration vector in the first sensed data and the reference acceleration vector, and issues a prompting message through the speaker in response to the earphone being worn at the incorrect position.

### Claim 6 (depends on 1)

The earphone of claim 1, wherein the data transformation model is adjusted based on an installed position of the inertial sensor.

### Claim 7 (depends on 1)

The earphone of claim 1, wherein the data transformation model is established by machine learning, deep learning or mathematical modeling.

### Claim 8

A system for transforming sensed data, the system comprising:
an inertial sensor sensing a motion of a user to generate first sensed data; and
a processor communicatively connected to the inertial sensor and pre-storing a data transformation model, wherein the data transformation model converts an acceleration vector and an angular velocity vector in the first sensed data into an acceleration vector and an angular velocity vector in second sensed data, and outputs the acceleration vector and the angular velocity vector in the second sensed data to an identification model pre-stored in the processor.

### Claim 9 (depends on 8)

The system of claim 8, wherein the processor determines whether the motion is correct by using the identification model and the second sensed data and generates an output result.

### Claim 10 (depends on 8)

The system of claim 8, wherein the identification model is updated according to the second sensed data.

### Claim 11 (depends on 8)

The system of claim 8, wherein the processor further pre-stores a reference acceleration vector, and updates one of the first sensed data and the second sensed data based on a vector angle between the acceleration vector in the first sensed data and the reference acceleration vector.

### Claim 12 (depends on 8)

The system of claim 8, wherein the inertial sensor and the processor are both disposed in a mobile device.

### Claim 13 (depends on 12)

The system of claim 12, wherein the mobile device further comprises a speaker coupled to the processor, the processor further pre-stores a reference acceleration vector, the processor determines that the mobile device is worn at an incorrect position based on a vector angle between the acceleration vector in the first sensed data and the reference acceleration vector, and the processor issues a prompting message through the speaker in response to the mobile device being worn at the incorrect position.

### Claim 14 (depends on 12)

The system of claim 12, wherein the data transformation model is adjusted based on an installed position of the inertial sensor in the mobile device.

### Claim 15 (depends on 8)

The system of claim 8, wherein the data transformation model is established by machine learning, deep learning or mathematical modeling.