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.

Register Log in

[Moved] Code for Gender Recognition using Auto-Correlation in Matlab

Status
Not open for further replies.
K

Kashyap Cherukuri

Junior Member level 3
Junior Member level 3
Joined
Jan 2, 2011
Messages
25
Helped
4
Reputation
8
Reaction score
3
Trophy points
1,293
Location
Kharagpur/Hyderabad
Activity points
1,419
This is another code that I have written for gender recognition in Matlab.
I have attached the files.
 

Attachments

  • gender_recognition.zip
    10.6 KB · Views: 1,388

Re: Code for Gender Recognition using Auto-Correlation in Matlab

Great code, you know your way around Matlab. It doesn't work on my machine though. It says "could not recognize. Try speaking slowly." But I think its my mic because the time-domain plot reveals a large transient at the beginning of each recording. I am going to keep looking through it when I get more time, because I also want to find out how autocorrelation is used... I don't understand how that can reveal gender.
 
Re: Code for Gender Recognition using Auto-Correlation in Matlab

The large transient at the beginning doesn't matter much. Autocorrelation was used to find out the fundamental frequency of the speech signal. The fundamental frequency of a male and female voice differs. The fundamental frequency of a male voice is typically between 80 and 155Hz while that of a female is between 165 and 255Hz. Try out this link Voice frequency - Wikipedia, the free encyclopedia
 
  • Like
Reactions: Lakshit and trav1s

    trav1s

    trav1s

    Points: 2
    Helpful Answer Positive Rating
    L

    Lakshit

    Points: 2
    Helpful Answer Positive Rating
    V

    Points: 2
    Helpful Answer Positive Rating
Re: Code for Gender Recognition using Auto-Correlation in Matlab

Ah ok I discovered if I just hum instead of talking it works much better. Maybe my natural speech just has too much pitch-modulation for it to handle.
 
  • Like
Reactions: alikh

    A

    alikh

    Points: 2
    Helpful Answer Positive Rating
Re: Code for Gender Recognition using Auto-Correlation in Matlab

Kashyap Cherukuri said:
This is another code that I have written for gender recognition in Matlab. I have attached the files.
View attachment 59385.
Click to expand...

I am currently doing a similar project and I really want to now the kind of approach that's used in here. I am doing this in my hardware interface course and using Atmel microcontroller programmed in Assembly language. I really really want to know what kind of approach along with hardware stuff you used here since last month I was looking for a way to reduce it's complexity but couldn't find.
Thanks and yes, your code looks great!
 

Re: Code for Gender Recognition using Auto-Correlation in Matlab

Yes trav1s. Humming gives a single tone. In other words humming is nothing but creating a sound with a single frequency while speaking produces sound over a spectrum of frequencies. So humming would give better results
 

Re: Code for Gender Recognition using Auto-Correlation in Matlab

thankss alot buddy...it works...awesome..:)

- - - Updated - - -

i need some modifications..i want to make a program in matlab that execute in three steps like i gave some sound having noise+music instrument and vocals..in the 1st step it remove noise then pass to the 2nd step where it filters the vocals and remove instruments and finally recognize the vocals whether its male or female...can u do it???
 

Re: Code for Gender Recognition using Auto-Correlation in Matlab

"could not recognize. Try speaking slowly."

it is not playing my voice i am doing it on my laptop.....................
 

Re: Code for Gender Recognition using Auto-Correlation in Matlab

hey this code is really awesome :) could you please tell me what had you used for feature extraction?
 
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top