Try SoX, that should do it.

 

There are no stupid questions.

The problems of matching a soundtrack to a given frame rate are at least as old as the pyramids. The problem is that you have to decide what frame rate to record at. (Actually, the problem is that there are historical differences between the standards that specify a particular frame rate. For example, 25 or 50 frames/second for PAL and 29.97 or 59.94 fps for NTSC.) Let's say I choose a frame rate of 25 fps for a recording. The movie is 90 minutes long. Now this movie is going to be released on Blu-ray. The frame rate there is 24 fps. The same movie now has a length of almost 94 minutes.

Calculation

90 minutes x 60 = 5400 seconds x 25 fps = 135.000 frames
135.000 frames / 24 fps =5625 seconds.

At 24 fps, our movie now has a runtime that is 3 minutes and 45 seconds longer! Our soundtrack is now asynchronous to the picture by these almost 4 minutes. The audio is initially synchronized with the picture. It becomes increasingly asynchronous as the running time progresses. This problem has nothing to do with a specific sample rate.

As this thread shows, a few already exist. However, the accuracy with which they do their work varies. Normally the offset to the image should not exceed 5 milliseconds (sound advanced with reference to vision) resp. 15 milliseconds (sound delay with reference to vision). Otherwise the offset between image and sound will be noticeable. It is therefore important that the conversion process is as accurate as possible.

This is not correct. You can use any sample rate with any frame rate. What can happen is that, for example, a specification requires you to use certain sample rates. For example, you cannot use a 44.1 kHz sample rate with Blu-ray. This applies to all frame rates, not just 24 fps.

 

Last edited: Sep 28, 2024

This is all really interesting, thanks for the detailed description. The final output medium is what is most important, whether it is Blu-ray or something else, in addition to the parameters of the client. Then the work is done accordingly. Did you end up trying SoX for SRC?

 

When dubbing, it is now common practice to make a separate recording for each frame rate required. Digital technology allows this to be done simultaneously.

I haven't gotten around to comparing the individual SRCs yet. Right now I'm still using RX from iZotope.