IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Converting my voice to target's voice
Warning
post Dec 19 2012, 02:50
Post #1





Group: Banned
Posts: 18
Joined: 22-January 12
Member No.: 96612



I already tried AV Voice changer diamond and it sucked. Is there an application where I can give it comparable input with my voice and it can successfully convert it to the voice of the target?

Say, you want your voice to sound like Britney Spears.

Any application that can do this effectively?
Go to the top of the page
+Quote Post
Ethan Winer
post Dec 19 2012, 18:13
Post #2





Group: Members
Posts: 248
Joined: 12-May 09
From: New Milford, CT
Member No.: 69730



QUOTE (Warning @ Dec 18 2012, 20:50) *
Say, you want your voice to sound like Britney Spears.


I can't help you with software, but I can tell you that much of what makes one person sound the way they do is inflection and how the voice quality (pitch and formants) change over time. Especially for singers as in your example.

--Ethan


--------------------
I believe in Truth, Justice, and the Scientific Method
Go to the top of the page
+Quote Post
DVDdoug
post Dec 19 2012, 18:19
Post #3





Group: Members
Posts: 2592
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



No. That kind of thing only exists in the movies. There is still a demand for impressionists and mimics. wink.gif

You can make a voice sound mechanical or artificial, or you can change it so that you can't recognize the person's voice. And, there are tools for changing/correcting pitch while keeping the same natural voice.
Go to the top of the page
+Quote Post
Warning
post Dec 19 2012, 20:25
Post #4





Group: Banned
Posts: 18
Joined: 22-January 12
Member No.: 96612



I understand things like inflection and mannerisms have to be imitated manually, but what's the best program that can morph all the rest of the qualities? AV Diamond shows Timbre Pitch and another slider to mitigate "robotic" qualities in the voice, but it doesn't even get those right. It did change my voice but to nothing near of the samples.

I'm surprised with the advent of spectrographic audio views that nothing can compare two voices accurately. If I wanted to sound like a 6-year-old, increasing my pitch will just make me sound like a high-pitched adult, nothing near a kid. That's because a kid's voice is both higher pitched and it has fewer frequency bands (do they call these harmonics?)

Also, I'm not trying to imitate a singer, Spears was just an example. I'm well aware that even if I had Spear's accurate voice, singing and doing it right is far different than talking.

What I'm trying to do is re-dub this censored film and fill all the voids.

edit: I would not mind reciting an hour worth of lines by an intended character to give the program much more input to work with.
Go to the top of the page
+Quote Post
DVDdoug
post Dec 19 2012, 22:45
Post #5





Group: Members
Posts: 2592
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



QUOTE
What I'm trying to do is re-dub this censored film and fill all the voids.
I've done censored a couple of movies... Here are some of the tricks I used:

- Reverse or Scramble the voice to make it unintelligible (GoldWave has a "Sensor/Gibberish" feature.)
- Cut & splice-in a different word/phrase from the same character.
- Repeat a word. For example, instead of "You @#$%", it says "You... You...).
- Chop-out the word(s) and the video, and re-splice with a quick audio/video crossfade (so, there are no gaps). Sometimes you can remove an entire phrase/sentance without loosing anything important.
- Simply mute the word(s). (It helps to replace the background sound with a similar selection.)

This post has been edited by DVDdoug: Dec 19 2012, 22:56
Go to the top of the page
+Quote Post
Warning
post Dec 19 2012, 23:08
Post #6





Group: Banned
Posts: 18
Joined: 22-January 12
Member No.: 96612



QUOTE (DVDdoug @ Dec 19 2012, 22:45) *
QUOTE
What I'm trying to do is re-dub this censored film and fill all the voids.
I've done censored a couple of movies... Here are some of the tricks I used:

- Reverse or Scramble the voice to make it unintelligible (GoldWave has a "Sensor/Gibberish" feature.)
- Cut & splice-in a different word/phrase from the same character.
- Repeat a word. For example, instead of "You @#$%", it says "You... You...).
- Chop-out the word(s) and the video, and re-splice with a quick audio/video crossfade (so, there are no gaps). Sometimes you can remove an entire phrase/sentance without loosing anything important.
- Simply mute the word(s). (It helps to replace the background sound with a similar selection.)


No idea what you're talking about.
Go to the top of the page
+Quote Post
benski
post Dec 20 2012, 00:25
Post #7


Winamp Developer


Group: Developer
Posts: 670
Joined: 17-July 05
From: Brooklyn, NY
Member No.: 23375



QUOTE (DVDdoug @ Dec 19 2012, 17:45) *
QUOTE
What I'm trying to do is re-dub this censored film and fill all the voids.
I've done censored a couple of movies... Here are some of the tricks I used:


I think he's trying to do the opposite of what you suggest.
Go to the top of the page
+Quote Post
DVDdoug
post Dec 20 2012, 00:47
Post #8





Group: Members
Posts: 2592
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



QUOTE
No idea what you're talking about.
I was trying to suggest some minimally annoying/destracting ways of censoring a soundtrack.

If Bensky is right and you want to un-censor the soundtrack, whatever you do, the results are likely to be more distracting/annoying than what you already have.

You could get someone to mimic & lip-sync the entire soundtrack, but you'd also have to replace the background/Foley and the music.

This post has been edited by DVDdoug: Dec 20 2012, 00:48
Go to the top of the page
+Quote Post
Warning
post Dec 20 2012, 03:22
Post #9





Group: Banned
Posts: 18
Joined: 22-January 12
Member No.: 96612



Background wont be a problem, I have the isolated background music. I can't mimic properly, I lack such skill, which is why I was asking if good voice changer programs existed. It's almost 2013 and I thought it would be feasible by now. I vaguely recall "UMSIL" making something like this but dropping support for mic input because they were afraid people would use it to impersonate celebrities.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 19th September 2014 - 17:44