Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

MFCC definition problem?

Status
Not open for further replies.

dashkil

Member level 2
Joined
Mar 3, 2012
Messages
45
Helped
1
Reputation
2
Reaction score
0
Trophy points
1,286
Activity points
1,620
i read through many papers related to (MFCC) Mel-frequency cepstrum coefficient: I have confusion about MFCC process:

1) Many says i have do FT (Fourier Transform) on log-spectrum to obtain cepstrum (as mentioned in wiki : definition of power cepstrum)

2) Many says i have do IFT (Inverse Fourier Transform) on log-spectrum to obtain cepstrum.

3) Many refers to DCT-ll (Discrete cosine Transform-2) on Mel-log-spectrum to obtain MFCC.

My question is which one is right: 1) FT or IFT?
2) If IFT is correct than why they are using DCT-ll instead of IDCT-ll; since DCT-ll is similar to DFT or FT.
hence IDFT or IFT should be similar to IDCT-ll.

Need explanation and ur precious opinion. Thanks in advance.
 

Typically 3-th point is correct. Notice that the DCT will be applied on the log of energy of filterbanks.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top