How to fix speedup with AviSynth?

[poisondeathray]

Oh, I just realized the ends (part1 and part3 in my example) need the frame rate change, not the middle (part2). I'll leave it up to the OP to adjust the code.

LOL I did the same thing

[VideoFanatic]

You end up mixing filters on the wrong section or double filtering

Or putting the pieces back together incorrectly:

trim(0, 20512). Tweak(Hue=-0, Sat=1.0, Bright=0, Cont=1.0, Coring=False) ++trim(20512,78229). Tweak(Hue=-0, Sat=1.0, Bright=0, Cont=1.0, Coring=False)

In the sample video I did this and it joined the talking segment and wrestling segment into 1 video without problem so how is it putting the pieces back together incorrectly?:

trim(0, 465). Tweak(Hue=-0, Sat=1.0, Bright=0, Cont=1.0, Coring=False) ++trim(466,694). Tweak(Hue=-0, Sat=1.0, Bright=0, Cont=1.0, Coring=False)

[jagabo]

In the sample video I did this and it joined the talking segment and wrestling segment into 1 video without problem so how is it putting the pieces back together incorrectly?:

trim(0, 465). Tweak(Hue=-0, Sat=1.0, Bright=0, Cont=1.0, Coring=False) ++trim(466,694). Tweak(Hue=-0, Sat=1.0, Bright=0, Cont=1.0, Coring=False)

Yes, your code worked. But it is only using one filter. It's going to get very ugly with the full QTGMC, MvTools, Reinterlace sequence. At least break that out into a separate function. Then you can call it with a much cleaner looking:

Trim(...).ChangeSpeed()++Trim(...)++Trim(...).ChangeSpeed()

Or if you have many sections:

Trim(...).ChangeSpeed()\
++Trim(...)\
++Trim(...).ChangeSpeed()\
++Trim(...)\
++Trim(...).ChangeSpeed()\
 etc.

[VideoFanatic]

Poisondeathray - I've got it working with the below code. Annoyingly it doesn't work in multi-threaded. Anyway, you said to slow down the audio you used: TimeStretch(rate=83.41666667) Soundout(). So do I just use that code in the parts that are playing too fast? How do I import the AC3 file so I can do the timestretch?

Main=Mpeg2Source("E:\2 = New\SpeedUp.d2v", CPU=6)  

Main
Trim(0,300)
McTemporalDenoise(settings="medium", interlaced=true)
AssumeTFF() QTGMC(Preset="Super Fast", sharpness=0.7)
Vinverse() RemoveSpots() AddGrainC(var=1.0,uvar=1.0) SeparateFields() SelectEvery(4,0,3) Weave() 
A=last

Main
Trim(301,465)
McTemporalDenoise(settings="medium", interlaced=true)
AssumeTFF() QTGMC(Preset="Super Fast", sharpness=0.7)
Vinverse() RemoveSpots() AddGrainC(var=1.0,uvar=1.0) SeparateFields() SelectEvery(4,0,3) Weave() 
B=last

Main
Trim(466,694)
McTemporalDenoise(settings="medium", interlaced=true)
AssumeFPS(25) 
AssumeTFF()  
QTGMC(Preset="Super Fast", sharpness=0.7) 
source=last  
super = source.MSuper(pel=2)  
backward_vec = MAnalyse(super, overlap=4, isb = true, search=3)  
forward_vec = MAnalyse(super, overlap=4, isb = false, search=3)  
source.MFlowFps(super, backward_vec, forward_vec, blend=false, num=60000, den=1001) 
AssumeTFF().SeparateFields().SelectEvery(4, 0, 3).Weave()

C=last

A++B++C

[poisondeathray]

I would join the audio separately. You can use various methods

eg
http://forum.videohelp.com/threads/213080-Join-AC3-files

ie. Take the audio from section C using the timestretch script and using Trim() to mark the start/end points, or use an audio editor to slow it down , join them with one of the methods in the thread above

To import audio only in an avs script you can use NicAC3Source() , or FFAudioSource() . If there is an audio delay , don't forget to add that (it should be written in the name)

To cut audio you can use delaycut


Alternatively , you can specify audio with each section in the script by using AudioDub() , but this means you re-encode the entire audio (lose quailty in all the sections instead of just 1)

[poisondeathray]

There are inconsistencies in your script. You don't have "." between some of the filter calls . I don't know if that was a copy/paste error

[VideoFanatic]

I didn't have a period because my scripts worked fine without them. I'm still a bit confused about what I'm supposed to be doing to only slow down the audio on 1 section of the video. I've got this so far. I get the following error "no audio found in clip"

Main=Mpeg2Source("E:\2 = New\SpeedUp.d2v", CPU=6)  
NicAC3Source("E:\2 = New\SpeedUp T80 2_0ch 256Kbps DELAY 0ms.ac3")

Main
Trim(0,300)
# McTemporalDenoise(settings="medium", interlaced=true)
AssumeTFF() QTGMC(Preset="Super Fast", sharpness=0.7)
Vinverse() RemoveSpots() AddGrainC(var=1.0,uvar=1.0) SeparateFields() SelectEvery(4,0,3) Weave() 
A=last

Main
Trim(301,465)
# McTemporalDenoise(settings="medium", interlaced=true)
AssumeTFF() QTGMC(Preset="Super Fast", sharpness=0.7)
Vinverse() RemoveSpots() AddGrainC(var=1.0,uvar=1.0) SeparateFields() SelectEvery(4,0,3) Weave() 
B=last

Main
Trim(466,694)
# McTemporalDenoise(settings="medium", interlaced=true)
AssumeFPS(25) 
AssumeTFF()  
QTGMC(Preset="Super Fast", sharpness=0.7) 
source=last  
super = source.MSuper(pel=2)  
backward_vec = MAnalyse(super, overlap=4, isb = true, search=3)  
forward_vec = MAnalyse(super, overlap=4, isb = false, search=3)  
source.MFlowFps(super, backward_vec, forward_vec, blend=false, num=60000, den=1001) 
AssumeTFF().SeparateFields().SelectEvery(4, 0, 3).Weave()
TimeStretch(rate=83.41666667) Soundout()
C=last

A++B++C

[jagabo]

You have to join the audio stream to the video stream. Use AudioDub().

[poisondeathray]

vid=Mpeg2Source("E:\2 = New\SpeedUp.d2v", CPU=6)   
aud=NicAC3Source("E:\2 = New\SpeedUp T80 2_0ch 256Kbps DELAY 0ms.ac3")
AudioDub(vid,aud)
Main=last
.
.
.

[manono]

You end up mixing filters on the wrong section or double filtering

Or putting the pieces back together incorrectly:

trim(0, 20512). Tweak(Hue=-0, Sat=1.0, Bright=0, Cont=1.0, Coring=False) ++trim(20512,78229). Tweak(Hue=-0, Sat=1.0, Bright=0, Cont=1.0, Coring=False)

In the sample video I did this and it joined the talking segment and wrestling segment into 1 video without problem so how is it putting the pieces back together incorrectly?

Note the bold-faced frame numbers. You ended one trim with one frame and began the next with the same one. You increased the total frame count by one, probably something you didn't intend. I suppose it was meant as an example since your Tweak settings do nothing, but still...

[VideoFanatic]

OK so this is my test example what I've got so far. Is that correct? When I load my script I see 3 SoundOut popups. Should I just save an AC3 file of every file? Then what do I do because I need to join them again?

Should I just use the TimeStretch(rate=83.41666667) Soundout() text on the parts that I want slowed down and Soundout() on parts that already play fine?

vid=Mpeg2Source("E:\2 = New\SpeedUp.d2v", CPU=6)   
aud=NicAC3Source("E:\2 = New\SpeedUp T80 2_0ch 256Kbps DELAY 0ms.ac3")
AudioDub(vid,aud)
Main=last

Main
Trim(0,300)
# McTemporalDenoise(settings="medium", interlaced=true)
AssumeTFF() QTGMC(Preset="Super Fast", sharpness=0.7)
Vinverse() RemoveSpots() AddGrainC(var=1.0,uvar=1.0) SeparateFields() SelectEvery(4,0,3) Weave() 
 Soundout()
A=last

Main
Trim(301,465)
# McTemporalDenoise(settings="medium", interlaced=true)
AssumeTFF() QTGMC(Preset="Super Fast", sharpness=0.7)
Vinverse() RemoveSpots() AddGrainC(var=1.0,uvar=1.0) SeparateFields() SelectEvery(4,0,3) Weave() 
TimeStretch(rate=83.41666667) Soundout()
B=last

Main
Trim(466,694)
# McTemporalDenoise(settings="medium", interlaced=true)
AssumeFPS(25) 
AssumeTFF()  
QTGMC(Preset="Super Fast", sharpness=0.7) 
source=last  
super = source.MSuper(pel=2)  
backward_vec = MAnalyse(super, overlap=4, isb = true, search=3)  
forward_vec = MAnalyse(super, overlap=4, isb = false, search=3)  
source.MFlowFps(super, backward_vec, forward_vec, blend=false, num=60000, den=1001) 
AssumeTFF().SeparateFields().SelectEvery(4, 0, 3).Weave()
TimeStretch(rate=83.41666667) Soundout()
C=last

A++B++C

[poisondeathray]

Take out all instances of soundout. Soundout is for encoding audio through avisynth

If you include audio in the script, you can feed the avs into an encoder that accepts avs scripts .

Or you can use 1 soundout() at the very end of the script, after A++B++C


Timestretch should only be used for part C, not part B (I'm assuming C is the only "bad" part)

[VideoFanatic]

I'm confused. In your early post you said "To slow down audio, you can use various audio editors e.g. audacity, timestretch in avisynth with soundout". So that's what I'm doing but now you're saying to not use SoundOut.

If I just use TimeStretch then it won't fix the audio. I thought the whole idea of this was to do both the video and audio fixes in the script? I'm aware that I'll probably have to re-encode the audio, that's fine but I just need to know what I'm supposed to do.

[poisondeathray]

Erase all instances of soundout , keep timestretch only for part C, add soundout() to the end after A++B++C . This will give you 1 audio (joined)

If you didn't want to use soundout to encode audio, you can comment out soundout and use that avs script as input into another audio encoder.

This is more damaging than if you cut the AC3, and only encode part C. Then join the original A and B

Does that make sense ?

[jagabo]

Why is SoundOut needed? I have a similar script and audio is passed through properly without using it (though it's only stereo).

There's another way to approach this, maybe a little more straight forward:

src = WhateverSource()
norm = FilterSequenceA(src)
slow = FilterSequenceB(src)

Trim(slow,...)++Trim(norm,...)++Trim(slow,...)

The only difficulty here is that you have to figure out what frame numbers in the slow video conform to the trims in the norm video.

[poisondeathray]

Why is SoundOut needed? I have a similar script and audio is passed through properly without using it (though it's only stereo).

soundout isn't required; it's an encoder

you can do the same thing without soundout , just feed the script into an encoder (without soundout)