Uncompressed AVI -> MKv With AviSynth

[royia]

Hello.
I have a project of our Home Video archive.

I captured all our videos (Hi8) into uncompressed AVI's (720 x 576, 25fps).

No all I want is to trim them (Each AVI contain few segments which I want to save independently), Crop, Deinterlace and remove noise and save as MKV (Encoding using x.264 and Lame MP3).

The question is how should I do it using scripts?

At first I tried MeGUI yet couldn't figure out what to do with the Audio (Since it's inside the AVI, Hence I get problems if I want to compress only part of the AVI).

Any advise how should I do it?

Thanks.

[Inspector.Gadget]

Since it's inside the AVI, Hence I get problems if I want to compress only part of the AVI

Not necessarily. I believe that you can use Avisynth's Trim() command in an audio-serving Avisynth script and achieve what you want.

[royia]

Not necessarily. I believe that you can use Avisynth's Trim() command in an audio-serving Avisynth script and achieve what you want.

I use MeGUI.
The Video input requires AVISynth script with no audio (audio = 'false'), Am I right?

So let's say I have an AVI file with audio and I want to trim it. Both the audio and the video.
I use trim command, Then what input should I give MeGUI in the Audio?

Thanks.

[poisondeathray]

no , you can use the same script for audio & video . Leave audio=true


No all I want is to trim them (Each AVI contain few segments which I want to save independently), Crop, Deinterlace and remove noise and save as MKV (Encoding using x.264 and Lame MP3).


deinterlace before you crop , because there are restrictions for cropping interlaced YV12 material

[royia]

no , you can use the same script for audio & video . Leave audio=true

Will MeGUI ignore the Video Processing command during the Audio Encoding process?

Thanks.

[poisondeathray]

not sure , but if your filters don't change framecount (e.g. not bob deinterlacing) , then you can use same script with cuts trim() without the video filters for the audio script, this way the cuts are in sync

or you can cut the uncompressed with vdub or other application

[royia]

OK, It worked.
Though for some reason the "Pre Processing" for the Audio Encode is long.

Thank You.

[tebasuna51]

Put your script for video and we can recommend the equivalent for audio.

[Umamio]

OK, It worked.
Though for some reason the "Pre Processing" for the Audio Encode is long.

If you leave "Normalize" checked in the megui audio settings preprocessing time is increased.

If you don't need it, uncheck it.
If you do need it, it's probably faster if you call Normalize() in the avisynth script itself rather than use the megui audio preprocessor.

[royia]

How do I use the Normalize() function on AVI file which has the audio built in?
In the reference - http://avisynth.org/mediawiki/Normalize - they always use it on wave files or channels extracted from the clip.

What if I apply it on the Clip itself? The function know to operate on Audio?

Thanks.

[Umamio]

I believe avisynth will automatically operate on the audio if you call Normalize() on a dubbed video clip.

If you want to normalise each audio channel separately you can still use the example in the documentation. You just need to put it right at the end of your script and modify the first line of the example so it becomes this

clip = last
left_ch = GetChannel(clip,1).Normalize()
right_ch = GetChannel(clip,2).Normalize()
audio = MergeChannels(left_ch, right_ch)
AudioDub(clip, audio)

[royia]

In my case it a mono stream.
So I'll just use Normalize().
I'll do that after Trim() for better performances.

Thanks.

[royia]

Anyhow, Posted a sample of the videos I'm dealing with:
http://extabit.com/file/28eej7i8kwnb5
I'd like to hear how experienced users would handle them (Deinterlacing, Noise Reduction, Handshakes Removel etc...).

[royia]

Posted a shorter (Different) file at:
http://www.sendspace.com/file/d6nd4p

This requires no waiting time and it is a fast site (60 MB).
I would be happy to know how the experienced user would deal with such material.
Thanks.

[Prettz]

If you leave "Normalize" checked in the megui audio settings preprocessing time is increased.

If you don't need it, uncheck it.
If you do need it, it's probably faster if you call Normalize() in the avisynth script itself rather than use the megui audio preprocessor.
Just an slightly off-topic question here: I assume megui uses all or nearly all the same audio libraries and methods as are present in BeSweet, and those are generally all the 'best of breed'. Is Avisynth's Normalize functionality as high quality as what megui uses? If you don't know the answer to that for sure, it's not a problem, I'm just curious.