- 两个关于降噪前后的语音SNR的计算参考
– SNR for audio .wav files and objective measures for evaluating filtering techniques
– How to calculate SNR of signals in MATLAB?
SNR计算
S1 = 原始干净语音
N1 = 噪声.
S2 = S1 + N1 (带噪语音)
S3 = 增强后的语音(使用某种算法降噪)
N2 = S3 - S1 (增强后语音中的残留噪声)
SNR的计算公式为(dB)
S
N
R
=
10
log
10
∥
s
i
g
n
a
l
∥
2
∥
n
o
i
s
e
∥
2
SNR = 10{\log _{10}}\frac{{{{\left\| {signal} \right\|}^2}}}{{{{\left\| {noise} \right\|}^2}}}
SNR=10log10∥noise∥2∥signal∥2
SDR的计算公式为
S
D
R
=
10
log
10
∥
X
c
∥
2
∥
X
−
X
c
∥
2
SDR = 10{\log _{10}}\frac{{{{\left\| {Xc} \right\|}^2}}}{{{{\left\| {X - Xc} \right\|}^2}}}
SDR=10log10∥X−Xc∥2∥Xc∥2
其中
X
c
X_c
Xc为带噪语音中的干净分量,
X
X
X为带噪语音,
X
c
−
X
X_c - X
Xc−X为带噪语音中的噪声分量。计算SNR提升量:
S
N
R
(
a
f
t
e
r
E
n
h
a
n
c
e
d
)
−
S
N
R
(
b
e
f
o
r
e
E
n
h
a
n
c
e
d
)
=
10
log
10
∥
S
1
∥
2
∥
N
2
∥
2
−
10
log
10
∥
S
1
∥
2
∥
N
1
∥
2
SNR(afterEnhanced) - SNR(beforeEnhanced) = 10{\log _{10}}\frac{{{{\left\| {S1} \right\|}^2}}}{{{{\left\| {N2} \right\|}^2}}} - 10\log 10\frac{{{{\left\| {S1} \right\|}^2}}}{{{{\left\| {N1} \right\|}^2}}}
SNR(afterEnhanced)−SNR(beforeEnhanced)=10log10∥N2∥2∥S1∥2−10log10∥N1∥2∥S1∥2
S
D
R
(
a
f
t
e
r
E
n
h
a
n
c
e
d
)
−
S
D
R
(
b
e
f
o
r
e
E
n
h
a
n
c
e
d
)
=
10
log
10
∥
S
1
∥
2
∥
S
3
−
S
1
∥
2
−
10
log
10
∥
S
1
∥
2
∥
S
2
−
S
1
∥
2
=
10
log
10
∥
S
1
∥
2
∥
N
2
∥
2
−
10
log
10
∥
S
1
∥
2
∥
N
1
∥
2
\begin{array}{l} SDR(afterEnhanced) - SDR(beforeEnhanced) = 10{\log _{10}}\frac{{{{\left\| {S1} \right\|}^2}}}{{{{\left\| {S3 - S1} \right\|}^2}}} - 10\log 10\frac{{{{\left\| {S1} \right\|}^2}}}{{{{\left\| {S2 - S1} \right\|}^2}}}\\ = 10{\log _{10}}\frac{{{{\left\| {S1} \right\|}^2}}}{{{{\left\| {N2} \right\|}^2}}} - 10\log 10\frac{{{{\left\| {S1} \right\|}^2}}}{{{{\left\| {N1} \right\|}^2}}} \end{array}
SDR(afterEnhanced)−SDR(beforeEnhanced)=10log10∥S3−S1∥2∥S1∥2−10log10∥S2−S1∥2∥S1∥2=10log10∥N2∥2∥S1∥2−10log10∥N1∥2∥S1∥2
从公式上看,两者完全相同。
SDR是(输入信号的功率)和(输入信号与增强信号之差的功率比),与SNR是一样的,在语音增强中,他们都反应了整体的性能。SDR的性能可以反应SNR的性能.
另外,在下面文献中也有类似的结论
Huang, Po-Sen, et al. “Joint optimization of masks and deep recurrent neural networks for monaural source separation.” IEEE/ACM Transactions on Audio, Speech, and Language Processing (TASLP) 23.12 (2015): 2136-2147.
SDR是输入信号的功率与输入信号与重构信号之差的功率之比。因此,SDR与经典的测量“信噪比”(SNR)完全相同,SDR反映了整体的分离性能。