Jump to content

Samplitude 8:Importing mp3 and remove audio


hanct
 Share

Recommended Posts

1)Whenever I try importing mp3 or watever format, samplitude 8 always have the habit of wanting to convert my mp3 to wave file...I dont want such conversion, is it possible?

2)I have a song with both music and singing voice. I want to remove the voice just to get the music itself, so as to form a karaoke song, which I can sing along with. Does samplitude able to do that? Doesnt matter if samplitude can do that perfectly or not..just wanna know if samplitude can do that.

3) I want to record my voice (together with a music). But my voice is horrible. I decided to use a analyser, to analyse a popular singer voice (whose voice is good), so that I can take the characteristic of that singer voice, and apply it to my voice. So that the result is that singer voice, but my lyrics... Does current audio technology allow such thing? Can samplitude do that?

Link to comment
Share on other sites

Guest Guest_Tekker
1)Whenever I try importing mp3 or watever format, samplitude 8 always have the habit of wanting to convert my mp3 to wave file...I dont want such conversion, is it possible?

The import feature always creates a separate wave file for that project. If you want to open a current file without the creation of a new wave file, then use the "Load Audio File" right off of the File menu - This is for version 8.... If you're using version 6, then go to File --> Open Project --> Wave.... I know it says open "wave" file, but it works with mp3's also. :huh: I don't know about v7, but it's probably similiar to one of these methods.

2)I have a song with both music and singing voice. I want to remove the voice just to get the music itself, so as to form a karaoke song, which I can sing along with. Does samplitude able to do that? Doesnt matter if samplitude can do that perfectly or not..just wanna know if samplitude can do that.

That depends... The way karaoke machines/programs work is they use a polarity reversal technique to remove anything panned in the "center", by flipping the polarity on one of the channels. There maybe some other EQ/filtering techniques but this center channel removal is the key to removing vocals. Unfortunately, this will also remove anything else that is panned in the center, like snare and kick drum, bass guitar, etc... There is no way to completely remove a vocal as they cover such a wide frequency range that you can't remove it completely without taking the rest of the music with it.

Now, certain songs will work better with this "polarity reversal" trick than others. If one song has everything panned straight down the center, then you can pretty much forget getting anything usable as all of the instruments are going to go with the vocals. But if there is a lot of wide stereo instruments (like guitar, synth, drums, etc) then those instruments, then you may be able to get by with it, because those instruments will be mostly in tact.

Another drawback to this method is reverb. If you remove a vocal that is panned in the center but it has a stereo reverb on it, then the vocal itself will be removed but you will still have the reverb "essence" of the vocal track. So you may not even be able to get rid of the vocals completely (this depends on how much reverb is on the vocals and how loud it is). Also, if this happens to several of the other instruments that were panned in the center, then you will wind up with a song that sounds like it was washed in reverb because the dry instruments are gone and all you can hear is the reverb from those instruments.

So, now that you know a little bit about this technique, YES Samplitude can do this... And probably better than other karaoke machines because you have complete control over frequencies (with EQ and the FFT filter). To remove the center channel, use the stereo enhancer in the object editor and turn it all the way to the right so that it is in "stereo enhanced" mode. This will completely remove the center channel. Now if you want that "super wide" stereo sound, you can leave it like that or you can check the invert phase button for only one side (either left or ride, it doesn't matter). Then you should hear it come back into "mono". (You'll have to try it to hear what I mean. lol)

After you have your completely phase reversed track, there is another trick you can do. First make a copy of that track by pressing and holding the shift and ctrl key, then grab the object and move it down to the next track (holding the ctrl key makes a copy of the object and pressing the shift key keeps the object from shifting left or right while you drag it). Now on this track reset the stereo knob by double clicking on it (it should go back to "stereo" mode) and uncheck the phase reversal check box. This track should play normally when you play it soloed....

Now open the object editor again and add an FX Insert (Samplitude effect) called "FFT Filter". Here you can use the red line like EQ to remove frequencies. Start all the way at the right side (high frequencies) and draw a line across the bottom to reduce the high frequencies (you'll get the idea once you mess around with it) and keep going until you hear the vocals completely go away. You should be left with only the very low bass from the song. Now go back to the highs again and start about halfway up and start drawing a line back towards the middle to add some high frequencies back in. Stop once you hear the vocals start to come back. If you hear any vocals then remove some of the high frequencies remove them. You will have to experiment with the levels of the high frequencies since there is no midrange you will probably get a nasty harsh "hiss" sound, so you probably won't want to bring them all the way back up to 0dB.

This track should leave you with just the bass frequencies and some of the high frequencies and no vocals (or anything else in the mid range)... Now close the FFT filter and the object editor and mute this 2nd track. Play the first track by itself (the one with the center channel removed) and listen to it for a few seconds, then unmute the second track and hear the difference with both playing together. Adding the second track should give you back the bass frequencies that was cancelled out from the polarity reversal and also add in some more high frequencies, giving a much better sound overall to the final product.

Keep in mind that this method won't be perfect (but absolutely no vocal removal program is) and the results will depend a lot on how the song was mixed (panning wise and the amount of reverb).... But this should get you close to what you want. :lol: The only ideal way to remove the vocals without affecting the other instruments is to have access to the original "un-mixed" tracks and mute the vocal track before mixdown, but we don't have that luxary unfortunately. :lol:

3) I want to record my voice (together with a music). But my voice is horrible. I decided to use a analyser, to analyse a popular singer voice (whose voice is good), so that I can take the characteristic of that singer voice, and apply it to my voice. So that the result is that singer voice, but my lyrics... Does current audio technology allow such thing? Can samplitude do that?

Man, I wish that was possible! I think who ever figures out how to do that is going to be filthy stickin' rich! LOL

There are a lot of things that are involved sound. You can't for instance, use EQ or any filter to make a guitar sound like a violin even though both are stringed instruments.... (Unless you use a synth, which is kind of a different topic. lol) So the short answer is no Samplitude can't do that, but no other program can either.

Hope this helps! And if you have any questions on the vocal removal technique, feel free to ask. ;)

-tkr

Link to comment
Share on other sites

Guest Guest
3) I want to record my voice (together with a music). But my voice is horrible. I decided to use a analyser, to analyse a popular singer voice (whose voice is good), so that I can take the characteristic of that singer voice, and apply it to my voice. So that the result is that singer voice, but my lyrics... Does current audio technology allow such thing? Can samplitude do that?

Man, I wish that was possible! I think who ever figures out how to do that is going to be filthy stickin' rich! LOL

There are a lot of things that are involved sound. You can't for instance, use EQ or any filter to make a guitar sound like a violin even though both are stringed instruments.... (Unless you use a synth, which is kind of a different topic. lol) So the short answer is no Samplitude can't do that, but no other program can either.

Should work wit SAM and the FFT Filter Analyse :huh:

Make a analyse of a voice and take the result as the frequencecurve

for a other file.

Regards

Link to comment
Share on other sites

Guest Guest

no,...it wont work.

FFT Filter Analyse is an EQ. It will only apply the same fft (eq sound) characteristic to the new file. What your asking is impossible. Not to mention,..an idiotic approach.

Be the true musician you once wanted to be and sing it your best,..or higher someone who can sing it.

Link to comment
Share on other sites

BUT...

you CAN get your voice to sound as though it was recorded with the same signal chain as the better sounding voice,...(relatively close) but the actual singing...NO.

Link to comment
Share on other sites

Guest Guest_Tekker
BUT...

you CAN get your voice to sound as though it was recorded with the same signal chain as the better sounding voice,...(relatively close) but the actual singing...NO.

There's a lot more to signal chain than just EQ, which can only boost or cut what is already present in the signal. If it was truely that easy, then no one would spend the money on the high end gear. :huh:

-tkr

Link to comment
Share on other sites

Guest Guest_Tekker
Thank you very much!!!! Ooh my gosh...U reply really a lot!!!! I love such reply! I gona try out what u say now....hopefully it will work. Thanks really a lot!!!!

I've been known to get a little long winded! Haha.... :huh:

Hopefully it will work out and give you the results you are looking for. :lol:

-tkr

Link to comment
Share on other sites

Guest I said "relatively close&qu
BUT...

you CAN get your voice to sound as though it was recorded with the same signal chain as the better sounding voice,...(relatively close) but the actual singing...NO.

There's a lot more to signal chain than just EQ, which can only boost or cut what is already present in the signal. If it was truely that easy, then no one would spend the money on the high end gear. :huh:

-tkr

There a multitude of IR's available on the net IR's from various high end tube equipment, mics and amps.. Not to mention those mic/amp/tube similator plugins. Antares Mic Modeler, Tube sim,.. etc.

Thats what I mean by relatively close,.. I never said to use "only EQ" for this and ...also never said it was "easy".

Unless you have the money for that all in one IR processing monster : Focusrite Liquid Channel :lol:

Link to comment
Share on other sites

Guest Guest_Tekker
There a multitude of IR's available on the net IR's from various high end tube equipment, mics and amps.. Not to mention those mic/amp/tube similator plugins. Antares Mic Modeler, Tube sim,.. etc.

But impulses work a bit differently than a straight EQ. I know the guys who made the Hydratone EQ made thousands of samples from the EQ units to create the impulses. This is more acurate to how that particular unit behaves and effects certain frequencies at different volumes, instead of taking one "snapshot" (so to speak) at a certain point in time and applying it over a song. I'm not even completely sure how the whole impulse thing works, and I've never heard the units they are sampling to hear how similiar/different they actually are either. The impulsing definitely sounds good (that I can say), but I can't say how close it is to the original.

Thats what I mean by relatively close,.. I never said to use "only EQ" for this and ...also never said it was "easy".

Weren't you the one who said...

"Should work wit SAM and the FFT Filter Analyse

Make a analyse of a voice and take the result as the frequencecurve

for a other file."

Because the FFT filter "is" an EQ, and I didn't see any other techniques mentioned.... So that's kind of how it sounded to me.... :huh:

-tkr

Link to comment
Share on other sites

Guest Guest
Weren't you the one who said...

"Should work wit SAM and the FFT Filter Analyse

Make a analyse of a voice and take the result as the frequencecurve

for a other file."

-tkr

Use Impulses for simulation of "better" equipment than what you have. (that's if your equipment isn't up to par) It will definately help your sound. Use individually on tracks (bass/guitars/voice etc.) and mixes. But can get overwhelming cuz you allway find yourself A/B'ing alot.

Samps fft isnt that great for copying fft (eq) curves onto another wave.

As far as karoke (removing vx) ,.. I have no idea.

Good luck.

Link to comment
Share on other sites

Guest Guest

oh..i meant to say:

Weren't you the one who said...

"Should work wit SAM and the FFT Filter Analyse

Make a analyse of a voice and take the result as the frequencecurve

for a other file."

-tkr

NO

Link to comment
Share on other sites

Guest Guest_Tekker
oh..i meant to say:
Weren't you the one who said...

"Should work wit SAM and the FFT Filter Analyse

Make a analyse of a voice and take the result as the frequencecurve

for a other file."

-tkr

NO

Oh, ok..... I think they need to remove the "guest" thing and make everyone register. Because you never know if the guest you are talking to is the same person. lol

Even though I have to post as a "guest" I at least use my "Tekker" screen name to make it easier to know who others are talking to. :lol:

I still can't figure out why I can't post with the Tekker user that I registered, but it won't let me post with it for some dumb reason. :huh:

-tkr

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
 Share

×
×
  • Create New...