# US 2014/0207460 A1 · Voice identification method and apparatus

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

- Number: US 2014/0207460 A1
- Document ID: US20140207460A1
- Status: Published application
- Kind: A1
- Application number: 14/100,822
- Filed: 2013-12-09
- Published: 2014-07-24
- Earliest claimed: 2013-01-24
- Assignees: Huawei Device Co., Ltd.
- Applicants: Huawei Device Co., Ltd.
- Inventors: Hongrui JIANG; Xiyong WANG; Junbin LIANG; Weijun ZHENG; Junyang ZHOU
- Web page: https://patentgrep.com/patent/US20140207460A1

## CPC

- G10L 17/20: Pattern transformations or operations aimed at increasing system robustness, e.g. against channel noise or different working conditions
- G10L 17/12: Score normalisation
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

Embodiments of the present invention provide a voice identification method, including: obtaining voice data; obtaining a first confidence value according to the voice data; obtaining a noise scenario according to the voice data; obtaining a second confidence value corresponding to the noise scenario according to the first confidence value; and if the second confidence value is greater than or equal to a pre-stored confidence threshold, processing the voice data. An apparatus is also provided. The method and apparatus that flexibly adjust the confidence value according to the noise scenario greatly improve a voice identification rate under a noise environment.

## Claims

### Claim 1

A voice identification method, comprising:
obtaining voice data;
obtaining a first confidence value according to the voice data;
obtaining a noise scenario according to the voice data;
obtaining a second confidence value corresponding to the noise scenario according to the first confidence value; and
processing the voice data if the second confidence value is greater than or equal to a pre-stored confidence threshold.

### Claim 2 (depends on 1)

The method according to claim 1, wherein the noise scenario comprises:
a noise type; and
a noise magnitude.

### Claim 3 (depends on 2)

The method according to claim 2,
wherein obtaining a noise scenario according to the voice data comprises:
obtaining, according to the voice data, a frequency cepstrum coefficient of a noise in the voice data; and
obtaining, according to the frequency cepstrum coefficient of the noise and a pre-established noise type model, the noise type of the voice data.

### Claim 4 (depends on 3)

The method according to claim 3, wherein the pre-established noise type model is established by a method comprising:
obtaining noise data;
obtaining a frequency cepstrum coefficient of the noise data according to the noise data;
processing the frequency cepstrum coefficient according to an EM algorithm; and
establishing the pre-established noise type model.

### Claim 5 (depends on 3)

The method according to claim 3, wherein the pre-established noise type model is a Gaussian mixture model.

### Claim 6 (depends on 2)

The method according to claim 2, wherein the obtaining a noise scenario according to the voice data comprises:
obtaining, according to the voice data, a feature parameter of the voice data;
performing voice activity detection according to the feature parameter; and
obtaining the noise magnitude according to a result of the voice activity detection.

### Claim 7 (depends on 6)

The method according to claim 6, wherein the noise magnitude comprises:
a signal-to-noise ratio; and
a noise energy level.

### Claim 8 (depends on 1)

The method according to claim 1, wherein obtaining a second confidence value corresponding to the noise scenario according to the first confidence value comprises:
obtaining, according to a correspondence between the noise scenario and pre-stored empirical data of a confidence value adjusting value, a confidence value adjusting value corresponding to the noise scenario; and
adjusting the first confidence value according to the confidence value adjusting value; and
obtaining the second confidence value;
wherein the adjusting includes one of the group consisting of: increasing, decreasing, and leaving unchanged.

### Claim 9 (depends on 1)

The method according to claim 1, further comprising:
prompting a user if the second confidence value is smaller than the confidence threshold.

### Claim 10

A voice identification apparatus, comprising:
an obtaining unit, configured to obtain voice data;
a first confidence value unit, configured to receive the voice data obtained by the obtaining unit and to obtain a first confidence value according to the voice data;
a noise scenario unit, configured to receive the voice data obtained by the obtaining unit and to obtain a noise scenario according to the voice data;
a second confidence value unit, configured to receive the noise scenario obtained by the noise scenario unit and the first confidence value obtained by the first confidence value unit and to obtain a second confidence value corresponding to the noise scenario according to the first confidence value; and
a processing unit, configured to receive the second confidence value obtained by the second confidence value unit and, if the second confidence value is greater than or equal to a pre-stored confidence threshold, to process the voice data.

### Claim 11 (depends on 10)

The apparatus according to claim 10, further comprising:
a model establishing unit, configured to obtain noise data, to obtain a frequency cepstrum coefficient of the noise data according to the noise data, to process the frequency cepstrum coefficient according to an EM algorithm, and to establish a noise type model.

### Claim 12 (depends on 11)

The apparatus according to claim 11, wherein the noise scenario unit comprises:
a noise type unit, configured to obtain, according to the voice data obtained by the obtaining unit, a frequency cepstrum coefficient of a noise in the voice data and to obtain, according to the frequency cepstrum coefficient of the noise and the noise type model established by the model establishing unit, a noise type of the voice data.

### Claim 13 (depends on 10)

The apparatus according to claim 10, wherein the noise scenario unit comprises:
a noise magnitude unit, configured to obtain, according to the voice data obtained by the obtaining unit, a feature parameter of the voice data, to perform voice activity detection according to the feature parameter, and to obtain a noise magnitude according to a result of the voice activity detection.

### Claim 14 (depends on 10)

The apparatus according claim 10, further comprising:
a storage unit, configured to store a confidence threshold and empirical data of a confidence value adjusting value.

### Claim 15 (depends on 14)

The apparatus according to claim 14, wherein the second confidence value unit is configured to:
obtain, according to a correspondence between the noise scenario and the empirical data of the confidence value adjusting value, a confidence value adjusting value corresponding to the noise scenario; and
adjust the first confidence value according to the confidence value adjusting value, wherein the adjusting includes of one of the group consisting of: increasing, decreasing, and leaving unchanged; and
obtain the second confidence value.

### Claim 16

A mobile terminal, comprising:
a processor; and
a microphone;
wherein the microphone is configured to obtain voice data; and
wherein the processor is configured to obtain a first confidence value according to the voice data, to obtain a noise scenario according to the voice data, to obtain a second confidence value corresponding to the noise scenario according to the first confidence value, and, if the second confidence value is greater than or equal to a pre-stored confidence threshold, to process the voice data.

### Claim 17 (depends on 16)

The mobile terminal according to claim 16, further comprising
a memory, configured to store empirical data of a confidence value adjusting value and the confidence threshold.

### Claim 18 (depends on 17)

The mobile terminal according to claim 17, wherein the processor is configured to:
obtain, according to correspondence between the noise scenario and the empirical data, a confidence value adjusting value corresponding to the noise scenario;
adjust the first confidence value according to the confidence value adjusting value;
obtain the second confidence value; and
process the voice data if the second confidence value is greater than or equal to the confidence threshold.
