Semitones Shift in NTSC to PAL

[Mr_Odwin]

I'm hoping someone can help me out!
I've been encoding an NTSC video and changing it into PAL format by speeding it up from 23.976 to 25 fps. I'd like to change the pitch of the sped-up audio so that it is similar to the original. I understand that this is only a semitone or two or something like that, but what is the exact figure? How many semitones do I have to lower the new audio? Is is 1, 2, or some crazy decimal?

I'll be using soundstretch and would really like to get it as close as it can be to the original pitch.

[XIII]

Hi

Why not just use besweet to change it from 23.976 to 25 fps
wont sound any different from the original.


XIII

[Mr_Odwin]

I used besweet and it sounds different. I found that using soundtouch in besweet lead to the vid and audio going out of sync but just using the ota thing works fine, sync-wise. I'd like to use soundstretch before besweet to sort out the pitch.

[XIII]

Hi

Why not just use the OTA in besweet on the original track ?

No need to do anything else.

XIII

[Mr_Odwin]

That's what I'm doing. It sounds different. I use this command:

"C:\Program Files\BeSweet\BeSweet.exe" -core( -input "C:\Documents and Settings\James\My Documents\Working Folder\Title1.wav" -output "C:\Documents and Settings\James\My Documents\Working Folder\Title1justspedup.mp2" ) -ota( -r 23976 25000 -hybridgain ) -ssrc( --rate 48000 ) -mp2enc( -m s -b 192 -e )

[Prodater64]

I used besweet and it sounds different. I found that using soundtouch in besweet lead to the vid and audio going out of sync but just using the ota thing works fine, sync-wise. I'd like to use soundstretch before besweet to sort out the pitch.


You can try several approachs.

1 - Managing sound through avisynth and encoding with, for example tmpgenc (better if you set its audio engine to external toolame)

http://www.avisynth.org/TimeStretch

In this case your audio output will be mp2.

2 - With any of my tools (and DIKO) using "DIKO only audio mode":

DIKO only audio (http://www.vmesquita.com/forum/index.php?topic=2951.0)
ProCalc ASPA Lite (http://forum.doom9.org/showthread.php?t=98907)
ProCalc ASPA (http://forum.doom9.org/showthread.php?t=96834)

With these tools you can process several audio files at once.

With DIKO you can do all (a/v, muxing, authoring) encoding job. "DIKO only audio" is a workaround from myself.
With this method your output could be mp2 (BeSweet) or ac3 (if you have Sonic Foundry Softencode software). Im not sure now, if DIKO manages 23.976 to 25 fps with assumefps(25), i think that yes, so it will be useful for you. But if DIKO does it with convertfps() or changefps(), the tempo will not change and then it would be useless for you.

4 - Do encode video using convertfps() or changefps() in your avs script. You don't need to process audio, only convert it to mp2 or ac3 if your target is (S)VCD or DVD. I don't know how about video playing smoothness. Test yourself.

5 - Do encode video resizing to PAL but "DON'T CHANGING THE FPS".
You will obtain a video stream, lets say 720*576 at 23.976 fps. Just apply DGPulldown 23.976 to 25 fps. You don't need to process audio. Again i don't know how about video playing smoothness. Test yourself.

[charleski]

It'll be a fraction of a semitone. If you think of it in terms of frequency: a 100Hz tone at 23.976fps will be 104.27Hz at 25fps (new f = 25*100/23.976). So your new audio will be 4.27% faster or at a frequency of 1.0427 x the original.
If you're using Audition, you can plug that number straight into the time stretch effect. If you need to work it out in semitones you need to do more maths. One octave is 2x the frequency of the one before it, and each semitone is 1.05946x the one before (the 12th root of 2). A frequency change of 1.0427 equates to 0.724 semitones by my calculations (12/(ln2/ln1.0427) and yes, I did have to refresh my maths a bit :).

[Mr_Odwin]

Thanks everyone. It's much appreciated, and I'll look into everything that's been said.

[XIII]

Hi

Why not use BeSweetGUI or BeLight instead of command line, makes things a lot easier.


XIII