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

seperation of simultaneous speech

Status
Not open for further replies.
U

urbanfreak

Newbie level 5
Newbie level 5
Joined
Jan 13, 2005
Messages
10
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
87
I have been working on seperating 2 voices recorded using a single microphone. Is it even possible? I have reached a dead end on the project.
Consider both the voices are male. Also, will it be possible if a sample is taken before hand of the two voices.

Please help!

Thanks

Anuj
 

thanks for that article. it was helpful. but no definite solution was provided.
would like to know if anyone has ever attempted this and succeeded.
 

This a nice project ..We humans are able to discriminate some information and just pay attention to the one important .I think that the approach is probably with neural networks
 

neural networks maybe can help you recognize.
but how about seperating...
i have no idea how to go about this..

but interesting project :)
 

I've read some papers about this topic. This topic is broad : Blind Source Separation. In general, people use more than 1 channel to separate mix signals. This is the "party cocktail problem". There're several methods but I prefer ICA (Independent Component Analysis).

I have read a paper recently, about sources separation using 1 channel only. That's really interested but I haven't tested with it . You can search with the keyword "Logmax algorithm". Hope you enjoy.
 

    U

    urbanfreak

    Points: 2
    Helpful Answer Positive Rating
thanks a lot tienpfiev. logmax algorithm was helpful.
more inputs are always welcome.

:)

anuj
 

consider usig neural network at least it is the way human can seperate voices
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top