Silly ME!!! I must be audio conversion challenged

[tonofsteel]

My first question is, is there any other way to change the fps for an audio part other than stretching or compressing it time wise??? I know its barely noticible but is there a way to do it without changing the pitch at all???

As well i have been following guides, and replies to other people that have asking this problem, but I cannot get my audio and video to sync after a avi pal to dvd ntsc conversion.

I got the video looking slick, and I extract the audio as a wav file ensuring that it is 48kHz using virtual dub. I then use besweet to do the ntsc to pal conversion for audio, outputting another wav file. I then use vegas 4's AC3 plugin to encode the wav to a ac3 file. When i mix that bunch of misfits together (the video and audio) I get mayhem. The audio runs away from the video at an exponential rate. And by time the middle of the movie comes around the audio is SCENES away from the video.

I have tried doing this several times using basically the same steps and get the same results each time. When i check the lengths between the original wav and the new wav, the new one is a few seconds longer, and this is right isnt it? So why would the audio run away from the video??

Does besweet just time warp the audio like goldwave? Or does it do some other kinky stuff to the wav file i dont know about? Should i ditch besweet and go to goldwave, or is there a way to get besweet working?

[jimmalenko]

I cannot get my audio and video to sync after a avi pal to dvd ntsc conversion.

I then use besweet to do the ntsc to pal conversion for audio, outputting another wav file.

Is it just me, or are you doing the wrong conversion ?

You start with PAL and transform it to NTSC, yet with the audio you are saying that the source is NTSC and to convert it to PAL ? No wonders it is scenes away.

[tonofsteel]

Stupid mistake, i am trying to convert a video from pal to ntsc. As well i am trying to convert the audio from pal to ntsc, just wrote it in the wrong order there, sorry.

[tonofsteel]

Once again I have guided myself out of my own problem. Why do all the guides on the subject of converting PAL to NTSC tell you that you have to mess around with the audio file??? Thats just stupid. No really it is just stupid.

Apparantly when you use the AssumeFPS() in avisynth you have to mess around with the audio because AssumeFPS does not do a true framerate conversion it only changes how the file is served, or the speed of the frames or something. I am still not sure because noone on this forum seems to think my questions warrant any kind of response.

Use ConvertFPS() and your problems are all magically solved. It apparantly does do a true frame rate conversion, but attempts to do it as smooth as possible to avoid flickering etc. In fact with the video I am working with ConvertFPS() actually shows a better video than AssumeFPS(). AssumeFPS() was causeing some nasty flickering in some scenes which was eliminated by using ConvertFPS(). And you know what the kicker is?? YOU DONT HAVE TO TOUCH YOUR WAV. I have not seen this posted anywhere, but all I did with the audio was extract it directly using virtual dub, put it through my AC3 encoder, authored it with my m2v and BAM! she's synced and ready to go.

[bugster]

To convert from PAL to NTSC, you have to change the framerate from PAL at 25fps to NTSC 29.976fps or NTSCfilm at 23.976fps. If you use ConvertFPS(), then in the 1st case new frames have to be generated, usually by repeating frames, and in the second case frames have to be deleted. In either of these cases you increase the risk of jerky movement in your video. If it has worked for you this time then well done for you but be aware that future efforts using this method may not be so succesful.

BTW, changing the framerate and shrinking the audio is the way FILM is transferred to PAL DVD. I have used this technique many times for NTSC(film) to PAL conversion without sync problems, you must be doing something wrong somewhere.

[tonofsteel]

Thank you bugster!

Ok but when you use assumeFPS() and you shrink the audio, does that not change the pitch and everything of the audio, so its not really perfectionist exact?

So I should be using the AssumeFPS method so then no frames are dropped or added, and it just changes the playback rate? And if i use AssumeFPS then I will need to change the audio length? But if you change the length of audio and the playback rate would that not counteract each other and null the pitch change of the audio???

[bugster]

You are correct that assumeFPS() will change the playback length and so the audio length will have to be adjusted. Normall this would also change the pitch but some software methods can compensate for this. It's not perfect but it works. In practice the change in pitch from 23.976fps to 25fps (or vide versa) is not really noticeable. It may be obvious if you listen to the two soundtracks one after the other but in isolation it doesn't make the audio sound 'high' (or low).