# US 2009/0157398 A1 · Method and apparatus for detecting noise

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

- Number: US 2009/0157398 A1
- Document ID: US20090157398A1
- Status: Published application
- Kind: A1
- Application number: 12/081,409
- Filed: 2008-04-15
- Published: 2009-06-18
- Earliest claimed: 2007-12-17
- Assignees: Samsung Electronics Co., Ltd.
- Inventors: Nam-hoon Kim; Jeong-mi Cho; Byung-hwan Kwak; Ick-sang Han; Yiogchun Huang
- Web page: https://patentgrep.com/patent/US20090157398A1

## CPC

- G10L 21/0208: Noise filtering
- G10L 25/18: the extracted parameters being spectral information of each sub-band
- G10L 25/78: Detection of presence or absence of voice signals (switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10)
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A method of and apparatus for detecting noise are provided. The method of detecting noise includes: receiving an input of a voice frame and converting the voice frame into a filter bank vector; converting the converted filter bank vector into band data; calculating a weight Gaussian mixture model (GMM) for each band by using the converted band data; and detecting noise in the voice frame based on the calculation result.

## Claims

### Claim 1

A method of detecting noise comprising:
receiving an input of a voice frame and converting the voice frame into a filter bank vector;
converting the converted filter bank vector into band data;
calculating a weight Gaussian mixture model (GMM) for each band by using the converted band data; and
detecting noise in the voice frame based on the calculation result.

### Claim 2

The method of claim-1, wherein in the calculating of the weight GMM for each band, the weight GMM for each band is calculated by applying a weight for the band to a GMM for the band which is trained in advance.

### Claim 3 (depends on 1)

The method of claim 1, wherein in the converting of the converted filter bank vector into band data, the filter bank vectors for the entire frequency bands of the voice frame are converted into data for respective bands.

### Claim 4 (depends on 1)

The method of claims 1, wherein the weight GMM for each band is calculated according to equation below:
L

(
O
|
Φ
)
=
∑
m
=
1
M

[
α


log


w
m
+
∑
n
=
1
N

{
log


c
mn
+
log


N
m

(
O
m
|
μ
mn
,
σ
mn
)
}
]
where, L(O|Φ) denotes a likelihood, M denotes a filter bank order, N denotes the number of mixtures, Cmn denotes a mixture weight for each band, μmn denotes a Gaussian mean for each band, σmn denotes a Gaussian distribution for each band, wmn denotes a band weight, and a denotes a band weight scaling factor.

### Claim 5 (depends on 2)

The method of claim 2, wherein the GMM for each band is trained by using predetermined voice data and label data.

### Claim 6 (depends on 5)

The method of claim 5, wherein the weight for each band is trained by using the trained GMM for the band, voice data and label data.

### Claim 7 (depends on 6)

The method of claim 6, wherein the weight for each band is calculated according to equation below:
O
k

(
t
)
=
{
1
,
if


O

(
t
)
=
O
k

(
t
)
0
,
otherwise


P

(
O
k
|
O
,
W
k
)
=
1
N

∑
n
=
1
N

O
k

(
t
)
where, Ok(t) denotes a training label at time t, O(t) denotes a band GMM label at time t, K denotes a class index, and N denotes the number of entire labels of class K.

### Claim 8 (depends on 1)

A computer readable recording medium having embodied thereon a computer program for executing the method of claim 1.

### Claim 9

An apparatus for detecting noise comprising:
a filter bank analysis unit receiving an input of a voice frame and converting the voice frame into a filter bank vector;
a band data converting unit converting the converted filter bank vector into band data;
a band weight GMM calculation unit calculating a weight GMM for each band by using the converted band data; and
a noise detection unit detecting noise in the voice frame based on the calculation result.

### Claim 10 (depends on 9)

The apparatus of claim 9, wherein the band weight GMM calculation unit calculates the weight GMM for each band by applying a weight for the band to a GMM for the band which is trained in advance.

### Claim 11 (depends on 9)

The apparatus of claim 9, wherein the band data converting unit converts the filter bank vectors for the entire frequency bands of the voice frame into data for respective bands.

### Claim 12 (depends on 9)

The apparatus of claim 9, wherein the weight GMM for each band is calculated according to equation below:
L

(
O
|
Φ
)
=
∑
m
=
1
M

[
α


log


w
m
+
∑
n
=
1
N

{
log


c
mn
+
log


N
m

(
O
m
|
μ
mn
,
σ
mn
)
}
]
where, L(O|Φ) denotes a likelihood, M denotes a filter bank order, N denotes the number of mixtures, Cmn denotes a mixture weight for each band, μmn denotes a Gaussian mean for each band, σmn denotes a Gaussian distribution for each band, wmn denotes a band weight, and a denotes a band weight scaling factor.