How does Tempo change work?

To many ads? Support ODJT and see no ads!

Kevin Kelly

New DJ
Aug 31, 2006
1,019
0
65
I know this is an incredibly stupid question but I cannot wrap my head around this...I know HOW it works but I'm just not sure why?

I'm using a unit now that has Tempo change and not just pitch bend and I wonder where it's been all my life! I never knew what I was missing! Can someone please explain to me - in laymens terms - how I can increase the tempo of a song yet the vocals don't start sounding like chipmunks? WTF????

:sqlaugh:
 
You need a unit with 'beatlock" . That way you can speed it up (or slow it down) without changing the key.
 
You press a button ;)
 
In essence, pitch bending is like controlling the speed of an LP... everything on the track is speeded up (or slowed down) depending upon one's needs. Push it up and the beat is faster... but the vocals sound like the Chipmunks.

Tempo bending affects the song tempo or beat without changing the pitch of the vocals. Very useful for beatmatching as the audience doesn't often notice the change.

True tempo bending can slow down (or speed up) the beat of a track while the vocals remain in the same pitch.

As to the dynamics, I can't help. Sorry. :)

Hope it helps.
 
In software, the simple answer is that samples of the song are either dropped (the song speeds up) or repeated (the song slows down).

It's implemented basically by converting the original audio sample stream to the desired duration by interpolating from the original audio samples.

This is overly simplistic, but is basically what happens.
 
In the easiest terms I can think of....

The speed is increased while the pitch is decreased

or

The speed is decreased while the pitch is increased


....instead of just the speed changing, like most "pitch" controls do on cheaper equipment.
 
Nope. It has nothing to do with raising ("increasing") or lowering ("decreasing") the pitch. The pitch remains unchanged because only the time domain is altered. You're thinking analog and not digital.
 
Explaining it in analog terms might be easier to comprehend though. If you speed up the record, the pitch will raise also, so if you counter the pitch change it would sound normal once again, just faster. ...right?
 
Hmmm. I don't quite agree with your analogy because it muddies the water further. Theoretically you're right, but in the real world it doesn't happen like that.

It's easier to think in terms of the sound being chopped up into samples (which is what happens in the digital domain). Then you either duplicate certain samples to stretch the time or remove them to shrink the time. By doing this you aren't changing the pitch data, so the pitch doesn't change.

Easy enough?
 
Huh??? :sqconfused: :sqconfused: :sqconfused:

Sorry, I'm more confused. Bryan, I think I understand a bit what you're talking about but I still don't understand why none of this would affect the vocals.
 
Kevin, it's because the pitch data (vocals have pitches, right?) doesn't change. You're not altering that data. You're simply duplicating some of the data (to make the song slow down) or deleting it (to make the song speed up).

The data you're repeating or removing doesn't change, so the pitch doesn't change.

Let's say that you sample the incoming music stream, and the actual pitch data value is 10. Let's also say for the sake of this example that the data value is constant over a one-second time frame. For our purposes, we'll sample that one-second frame five times. We get:

10 | 10 | 10 | 10 | 10 <- five samples of the value 10.

Now we want to slow down the music. To do that, we simply duplicate the value:

10 | 10 | 10 | 10 | 10 | 10 | 10 | 10 <- eight samples of the value 10.

The pitch data value, 10, didn't change, but we have more samples to output into the stream. We haven't changed the rate of samples at the output (five samples, remember?), so we have now stretched the song. We have slowed down the song without changing the pitch.

Now let's speed the song up. We take our original value:

10 | 10 | 10 | 10 | 10 <- five samples of the value 10.

And we remove some samples:

10 | 10 | 10 <- three samples of the value 10.

The pitch data value, 10, didn't change, but now we have fewer samples to output into the stream. We haven't changed the rate of samples at the output (five samples, remember?), so we have now shrunk the song. We have effectively sped up the song without changing the pitch.
 
Ok, that makes total sense! Thanks Bryan...I learned something today...just like my momma told me to do! LOL...

So is there any loss of audio quality when the data is deleted in order to speed up the song?
 
... So is there any loss of audio quality when the data is deleted in order to speed up the song?

I haven't been able to detect any deterioration in sound quality, even when bent to the extremes. When I practice with the 'phones it sounds fine.
 
So is there any loss of audio quality when the data is deleted in order to speed up the song?
Of course there is, otherwise every software that claims to have this ability would sound perfect when a song is sped up or slowed down! Everyone would have it. But everyone doesn't. It's difficult to do to perfection.

OtsDJ 1.75 does this almost perfectly. I say almost because when a song is slowed down drastically -- to a standstill -- there still is a metallic bathtub echo because of aliasing of the repeated sample boundaries.

I'll put a song snippet on my website showing OtsDJ's time-stretching capability and give link to it. How about "Play That Funky Music?"
 
Thanks, guys. Glad I could help.

FYI: I just redid the sample to stop at various parts of the song to show you how the OtsDJ time-scaling algorithms are free of audio artifacts. It's 1:30 long, and you may have to clear your browser's cache to get it.