本研究旨在對梅爾倒頻譜與一般倒頻譜進行比較。
優點：
Y
[
m
]
=
log
(
∑
k
=
f
m
−
1
f
m
+
1
|
X
[
k
]
|
2
B
m
[
k
]
)
{\displaystyle Y[m]=\log \left(\sum _{k=f_{m-1}}^{f_{m+1}}\left|X[k]\right|^{2}B_{m}[k]\right)}
|
X
[
k
]
|
2
{\displaystyle \left|X[k]\right|^{2}}
Y
[
m
]
=
log
(
∑
k
=
f
m
−
1
f
m
+
1
|
X
[
k
]
|
2
B
m
[
k
]
)
{\displaystyle Y[m]=\log \left(\sum _{k=f_{m-1}}^{f_{m+1}}\left|X[k]\right|^{2}B_{m}[k]\right)}
∑
k
=
f
m
−
1
f
m
+
1
|
X
[
k
]
|
2
B
m
[
k
]
{\displaystyle \sum _{k=f_{m-1}}^{f_{m+1}}\left|X[k]\right|^{2}B_{m}[k]}
X
[
k
]
{\displaystyle X[k]}
B
m
[
k
]
{\displaystyle B_{m}[k]}
f
1
,
f
2
,
f
3
,
.
.
.
{\displaystyle f_{1},f_{2},f_{3},...}
倒頻譜#特性_2 ，
[ 1]
↑ Jian-Jiun Ding, Advanced Digital Signal Processing class note, the Department of Electrical Engineering, National Taiwan University (NTU), Taipei, Taiwan, 2018.
![{\displaystyle Y[m]=\log \left(\sum _{k=f_{m-1}}^{f_{m+1}}\left|X[k]\right|^{2}B_{m}[k]\right)}](https://wikimedia.org/api/rest_v1/media/math/render/svg/94721c49f5b0d7a24db7a65507fa809eba4d77bd)
![{\displaystyle \left|X[k]\right|^{2}}](https://wikimedia.org/api/rest_v1/media/math/render/svg/f2f5b3e8988330a05df365228a2c3fecf1e77b16)
![{\displaystyle \sum _{k=f_{m-1}}^{f_{m+1}}\left|X[k]\right|^{2}B_{m}[k]}](https://wikimedia.org/api/rest_v1/media/math/render/svg/e469b00bb121cc503928d58114e02f0580510a62)
![{\displaystyle X[k]}](https://wikimedia.org/api/rest_v1/media/math/render/svg/b44d31b5739397a3a3a58892804677840e85bb2a)
![{\displaystyle B_{m}[k]}](https://wikimedia.org/api/rest_v1/media/math/render/svg/a08774ad3bcfc641065d34fe5c3d17a37448d2b0)
