|
|
|
11-05-2017, 09:24 PM
|
#41
|
Human being with feelings
Join Date: May 2007
Posts: 499
|
Thanks for this script. This thing is so cool!
|
|
|
01-12-2018, 10:40 AM
|
#42
|
Human being with feelings
Join Date: May 2010
Location: Norway
Posts: 7,318
|
Such a brilliant script.
Sometimes it worked perfect,
sometimes just 1/6 takes
and sometimes not getting any results.
But, then conditions were far from ideal I guess
Thank you!
__________________
Reaper x64, win 11
Composer, text-writer, producer
Bandcamp
|
|
|
02-19-2018, 04:11 AM
|
#43
|
Human being with feelings
Join Date: Jul 2008
Location: Athens / Greece
Posts: 625
|
Thank you so much for making this! Very cool!
|
|
|
04-30-2018, 05:34 AM
|
#44
|
Human being with feelings
Join Date: Apr 2016
Posts: 68
|
Is it possible to make script align items?
I mean.. I sometimes have to mix voice overs, and I have like 100 mastered orginal clips on 1st track, and 154 recorded on secound track. The problem is this 154 clips, are always 100 sentences, like in orginal, but sometimes it's cutted (without breaths) or serval clips are not in time (mismatched), or something. Now i spend a lot of time to only align this to orginal. Maybe I can do this automatically? waveforms are quite simmilar
|
|
|
05-01-2018, 12:04 AM
|
#45
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by xinky
Is it possible to make script align items?
I mean.. I sometimes have to mix voice overs, and I have like 100 mastered orginal clips on 1st track, and 154 recorded on secound track. The problem is this 154 clips, are always 100 sentences, like in orginal, but sometimes it's cutted (without breaths) or serval clips are not in time (mismatched), or something. Now i spend a lot of time to only align this to orginal. Maybe I can do this automatically? waveforms are quite simmilar
|
No, it is not possible for now.
As a woraround you can put each of editing tracks into subproject (one by one):
Select track > Action List > Track: Move tracks to subproject
|
|
|
05-02-2018, 05:41 AM
|
#46
|
Human being with feelings
Join Date: Apr 2013
Location: France
Posts: 9,875
|
@Xinky
I wonder if a tool like PluralEyes could work for you. It is designed for video but maybe it can simply process audio ? It aims to sync various files based on audio and can work with long files and different audio quality (different mic etc)
|
|
|
05-15-2018, 09:46 PM
|
#47
|
Human being with feelings
Join Date: Jul 2009
Posts: 429
|
In vocal aligning the most important thing for me comes in respecting sibilants and plosives. If the algorithm had a way to detect, align and preserve such in a non destructive manner, would be awesome.
Thanks for a very useful item.
|
|
|
07-19-2018, 06:16 PM
|
#48
|
Human being with feelings
Join Date: Nov 2017
Posts: 159
|
How do I install this/ReaPack on a Mac? I have been trying to find it but I can't seem to locate it.
|
|
|
07-19-2018, 06:19 PM
|
#49
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by boydwgrossii
How do I install this/ReaPack on a Mac? I have been trying to find it but I can't seem to locate it.
|
reapack.com
It is pretty hard to NOT find it.
|
|
|
07-22-2018, 02:01 AM
|
#50
|
Human being with feelings
Join Date: Dec 2015
Posts: 1,739
|
|
|
|
03-05-2019, 08:59 PM
|
#51
|
Human being with feelings
Join Date: Feb 2007
Location: South Carolina
Posts: 395
|
Thank you MPL, I have just got the script. Much appreciated
Video from Antonio was very helpful as well
|
|
|
05-05-2019, 03:25 AM
|
#52
|
Human being with feelings
Join Date: Feb 2019
Location: Southern Vermont
Posts: 864
|
Great idea
Brilliant idea for a script, can't wait to try this out on overdubbed vocals etc...
|
|
|
06-10-2019, 09:09 PM
|
#53
|
Human being with feelings
Join Date: Apr 2018
Posts: 67
|
Quote:
Originally Posted by mpl
reapack.com
It is pretty hard to NOT find it.
|
Hey man this is such an awesome tool thank you, 1 question though, once I apply the the stretch markers from the horizontal fader, can I re-visit the settings & pick from where I left off the sldier, thanks :')
|
|
|
06-10-2019, 11:32 PM
|
#54
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by pranavakazack
Hey man this is such an awesome tool thank you, 1 question though, once I apply the the stretch markers from the horizontal fader, can I re-visit the settings & pick from where I left off the sldier, thanks :')
|
Hmm you can save recall presets. And also when you press 'Get takes', it remove existing stretch markers from selected takes.
|
|
|
06-11-2019, 03:50 AM
|
#55
|
Human being with feelings
Join Date: Apr 2018
Posts: 67
|
Quote:
Originally Posted by mpl
Hmm you can save recall presets. And also when you press 'Get takes', it remove existing stretch markers from selected takes.
|
Oh okay, if it removes existing ones that's cool, I am new to Reaper, can you give me links me to cool scripts/actions which are like creative & fun to experiment around with like which make a sonic difference instead of an organizational one?
|
|
|
08-30-2019, 12:21 AM
|
#57
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Hopefully I`ll release soon second version with some major algorithm improvements. Any donation welcome and makes things work better (= I can spend more time on this).
|
|
|
08-30-2019, 02:41 AM
|
#58
|
Human being with feelings
Join Date: Nov 2014
Location: Can Tho - Viet Nam
Posts: 305
|
Quote:
Originally Posted by mpl
Hopefully I`ll release soon second version with some major algorithm improvements. Any donation welcome and makes things work better (= I can spend more time on this).
|
Oh ! So Great! I am looking forward to this
Thank you!
|
|
|
02-27-2020, 09:39 PM
|
#59
|
Human being with feelings
Join Date: Nov 2014
Location: Can Tho - Viet Nam
Posts: 305
|
mpl Align Takes has been updated today, thanks
|
|
|
02-27-2020, 10:20 PM
|
#60
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by lachinhan
mpl Align Takes has been updated today, thanks
|
Only for blind people. It's not a major update.
|
|
|
10-19-2020, 03:08 AM
|
#61
|
Human being with feelings
Join Date: Nov 2016
Posts: 226
|
Just came across this. This sounds quite promising. I was wondering what you think about my use case: I am recording solo guitar performances on steel string acoustic guitar. One recording will be in a room with good acoustics. The other recording will be in a room where I am shooting video, and will also capture audio there.
When I perform these pieces I slow down/speed up a bit. The differences in execution are not huge, but there are differences. Also, I play with a pick, and so transients should be pretty strong.
So what I need to do is fit my desired audio recording to the audio from the video, so that I have audio very well sync'd to the video.
Do you think this tool will work?
If it does I will buy you a drink next time I am in Moscow (I am there about 4 times a year)!
thanks,Brian
|
|
|
10-19-2020, 03:27 AM
|
#62
|
Human being with feelings
Join Date: Mar 2008
Location: Sydney, Australia
Posts: 3,955
|
wow, how did i miss this??
posting so it stays in my history until i get a chance to check it out
|
|
|
01-31-2021, 06:24 AM
|
#63
|
Human being with feelings
Join Date: Mar 2007
Location: Mediterrenean Sea
Posts: 977
|
Thank you so much for the script. This is a big time-saver
Donated!
|
|
|
01-31-2021, 08:14 AM
|
#64
|
Human being with feelings
Join Date: Aug 2011
Posts: 517
|
Wicked script!
|
|
|
02-06-2021, 08:30 PM
|
#65
|
Human being with feelings
Join Date: Mar 2007
Location: Mediterrenean Sea
Posts: 977
|
I,v been trying your script to align vocals and it is very promising.
I think it would be very welcome this script to have an option to add the stretch markers at zero crossing points and also set the items propierties to ,tonal-optimized,. How it is now it triggers a little ,pop, sound on some markers.
Big thanks
Last edited by antiClick; 02-06-2021 at 08:47 PM.
|
|
|
06-01-2021, 08:03 PM
|
#66
|
Human being with feelings
Join Date: Nov 2020
Location: Netherlands/Turkey
Posts: 36
|
I've tried the script but sadly it is not usable for me.
I had 3 takes which contained 2 words in 2 takes that where not aligned, and for some reason the script didn't put a stretch marker near those 2 words until I tweaked so many confusing knobs that my takes got loaded with a lot of markers. And what then happens is, it ruins the good parts. I wish it could detect words without pause/silence between them but it is not doing that in my situation.
After trying to tweak it for one hour I decided to stop with it.
It would have saved me a lot of time if this would work for me because I use a lot of vocal takes but sadly I need to align them manually.
Maybe it is useful if you sing with almost no difference to make it even more aligned.
Or maybe I can use the script for making starting points, but then again it doesn't put the markers on the right places so I still need to place them manually.
I have big respect for you that you made something like this even it is not usable for me.
I hope that I explained my issue without offending you.
Greetings and sry for my bad english
|
|
|
06-02-2021, 03:27 AM
|
#67
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by Mevzu
I've tried the script but sadly it is not usable for me.
I had 3 takes which contained 2 words in 2 takes that where not aligned, and for some reason the script didn't put a stretch marker near those 2 words until I tweaked so many confusing knobs that my takes got loaded with a lot of markers. And what then happens is, it ruins the good parts. I wish it could detect words without pause/silence between them but it is not doing that in my situation.
After trying to tweak it for one hour I decided to stop with it.
It would have saved me a lot of time if this would work for me because I use a lot of vocal takes but sadly I need to align them manually.
Maybe it is useful if you sing with almost no difference to make it even more aligned.
Or maybe I can use the script for making starting points, but then again it doesn't put the markers on the right places so I still need to place them manually.
I have big respect for you that you made something like this even it is not usable for me.
I hope that I explained my issue without offending you.
Greetings and sry for my bad english
|
yeah things like that can happen, the code is pretty rough, this was written 5 years ago, hopefully I`ll rebuild it from scratch someday. Or someone more clever will write it in a usable way.
|
|
|
01-07-2022, 11:58 AM
|
#68
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
2.0 version is out now.
Lots of stuff here so I quote myself:
Quote:
+ Complete core rebuild
+ Complete GUI rebuild
+ Internal cleanup
+ Core/Preset: rebuild preset system
+ Core/Preset: action to reset all settings
+ Core/Preset: right click on parameter reset parameter
+ Core/Preset: general preset management - Save, Save as new, Rename, Delete
+ Core/Global: allow to get reference take on initialization
+ Core/Global: allow to get dub take on initialization
+ Core/Global: if both ref and dub init check is on, get dub take starting from second take
+ Core/Global: allow to apply settings (get ref/dub takes and calculate best fit) on configuration change
+ Core/Global: option to clear stretch markers for dub takes on initialization
+ Core/Global: option to get time selection as data edges
+ Core/Global: option to init UI at mouse position
+ Core/Audio data: separate audio data settings
+ Core/Audio data: support takes with non-1x takerate
+ Core/Audio data: support looped takes
+ Core/Audio data: support dub edges outside reference take edges
+ Core/Audio data: get reference take parent track accessor data rather than take accessor
+ Core/Audio data: get dub takes source track track accessor data enclosed with reference accessor edges
+ Core/Audio data: remove FFT support, use band splitting based on JSFX 4-Band Splitter by Michael Gruhn
+ Core/Audio data: option to limit audio for analyze
+ Core/PeakFollower: separate peak follower (envelope creation) settings
+ Core/PeakFollower: allow to change window ovelap (improve envelope accuracy without changing window)
+ Core/PeakFollower: use pass counter for smoothing envelope
+ Core/PeakFollower: option to reduce points back to non-overlapped (overlap divider = 1) count
+ Core/MarkerGenerator: separate marker generator (phrase slicer) settings
+ Core/MarkerGenerator: remove rise/fall value, use relation value/RMS in defined area of points
+ Core/MatchAlgorithm: separate audio matching algorithm
+ Core/MatchAlgorithm: optionally disable on-the-fly stretching of dub array
+ Core/MatchAlgorithm: replace search area parameter with 'Brutforce search area', this is quantized to window/overlap length
+ Core/MatchAlgorithm: skip difference check at 0 values
+ Core/TakeOutput: separate take output settings
+ Core/TakeOutput: allow to change pitchshift/timestretch mode
+ Core/TakeOutput: allow to change stretchmarker fade size
+ GUI: allow to get reference and dub separately
+ GUI: retina scaling support
+ GUI: extending window extend UI
+ GUI: show all available dub takes
+ GUI: store window position
+ GUI: store window width/heigth
+ GUI: store window docker
+ GUI: scroll parameters list
+ GUI/OneKnobMode: reduce all UI to one knob if script width/height less than 500/200px (or more if retina)
+ GUI/OneKnobMode: override 'get reference/dub on init' checks
+ GUI/Shortcuts: Left / Right arrow move main knob
+ GUI/Shortcuts: G/g get ref/dub
+ GUI/Shortcuts: R/r get ref
+ GUI/Shortcuts: D/d get dub
+ GUI/Shortcuts: space trigger transport
+ GUI: show src/dest points with green/red colors
# remove nonGUi version, use OneKnobMode mode instead
# remove SWS dependency
# require VF_CheckFunctions 2.71+
|
This is first huge update since 2016. Some minor tweaks are coming soon.
This version requires VariousFunctions package which is paid.
You can still use last free version (don`t update it or revert back using ReaPack versioning system).
For more info, see here: https://forum.cockos.com/showpost.ph...30&postcount=1
Last edited by mpl; 01-07-2022 at 12:30 PM.
|
|
|
01-07-2022, 07:09 PM
|
#69
|
Human being with feelings
Join Date: Feb 2014
Posts: 310
|
Quote:
Originally Posted by mpl
This version requires VariousFunctions package which is paid.
You can still use last free version (don`t update it or revert back using ReaPack versioning system).
|
I completely understand putting content behnd a paywall (especially when a lot of other Reaper stuff is given away for free), but it would be nice if bugfixes and optimisations were added to the free version [preferably including retina support], even if the newer features are left out of it.
|
|
|
01-07-2022, 08:48 PM
|
#70
|
Human being with feelings
Join Date: Nov 2014
Location: Can Tho - Viet Nam
Posts: 305
|
I just donated for this. Hope to receive your activation code soon
Thanks
|
|
|
01-07-2022, 08:50 PM
|
#71
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by JayJSE2
I completely understand putting content behnd a paywall (especially when a lot of other Reaper stuff is given away for free), but it would be nice if bugfixes and optimisations were added to the free version [preferably including retina support], even if the newer features are left out of it.
|
For some scripts (retrospective record for example) I continue support both free and paid versions in parallel. For AlignTakes script is is pretty hard to fix old code (it was written pretty dirty). Anyway, I understand, a lot of people using this script, so I'll probably add version doing only data processing with optimized parameters and without UI (almost all audio related functions are human readable and placed inside main script).
EDIT: I just fixed hard dependency from v2, so now if you didn`t purchased, you will still be able to use full functional script. It will only trigger purchasing window at the script start (maybe I`ll make a Cockos-style 5-seconds purchasing window showing once, when triggered some of my scripts).
Last edited by mpl; 01-07-2022 at 09:28 PM.
|
|
|
01-07-2022, 09:44 PM
|
#72
|
Human being with feelings
Join Date: Nov 2014
Location: Can Tho - Viet Nam
Posts: 305
|
Version 2 seems more confusing and difficult to use than the old version. Because I don't understand it well, I feel it's not as effective as the previous version.
|
|
|
01-07-2022, 10:07 PM
|
#73
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by lachinhan
Version 2 seems more confusing and difficult to use than the old version. Because I don't understand it well, I feel it's not as effective as the previous version.
|
Re: VariousFunctions.
I removed all dependencies, so it should work even without purchasing (but with some settings limitations).
What you don`t understand? I know some settings aren`t really first-sight-understandable, it was made by user requests rather than for myself, so they are tech-related and not designed to be user-friendly. Probably I should add some well-optimized presets from the box?
So in short:
1) script reads audio data
2) then script search phrase/syllables. Phrase separators are the green points (future stretch markers). More points = better fit, but if you allow to stretch phrase too far, it can produce worse result.
3) then script compare audio data of dub takes to the reference take data. It follow 2-phrases blocks of data through dub take and try to move their separator the place where the whole block are best fit with reference audio data, just like what we usually do by hand. So the red points is where script found best fit for the green separator.
4) apply this best fit as a stretch markers
Ideally after all the setup it is enough to just trigger the script an move single knob:
Last edited by mpl; 01-07-2022 at 10:22 PM.
|
|
|
01-07-2022, 10:17 PM
|
#74
|
Human being with feelings
Join Date: Feb 2019
Posts: 586
|
Possible to get some audio examples? I would pay now but would like to try/hear it first. Tried to use with VariousFunctions 1.31 but it wouldn't open, is there a way to get that to work as a test drive?
|
|
|
01-07-2022, 10:26 PM
|
#75
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by fakemaxwell
Possible to get some audio examples? I would pay now but would like to try/hear it first. Tried to use with VariousFunctions 1.31 but it wouldn't open, is there a way to get that to work as a test drive?
|
Yes, you can try it.
Install latest VariousFunctions package (you can revert it back after try). You`ll see a purchase window once a REAPER start though.
Sorry, I`m not a major developer, there willn`t examples, colored pdf tutorials, teamviewer support etc, I just write code that do what it is expected to do. Any suggestions welcome though.
|
|
|
01-07-2022, 10:51 PM
|
#76
|
Human being with feelings
Join Date: Feb 2019
Posts: 586
|
I'm getting this error before Align Takes crashes when I first try to start it:
Quote:
mpl_Align takes.lua:703: attempt to call a nil value (global 'VF2_FormatToNormValue')
|
|
|
|
01-07-2022, 11:06 PM
|
#77
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by fakemaxwell
I'm getting this error before Align Takes crashes when I first try to start it:
|
Please update both AlignTakes and VariousFunctions packages. (Normally it doesn`t happen if you have enabled syncronization of whole ReaPack repository rather per-script sync)
Side note: unfortunately ReaPack doesn`t support dependencies, so thats why I have to link VariousFunctions package in such awful way.
|
|
|
01-09-2022, 02:02 AM
|
#78
|
Human being with feelings
Join Date: Mar 2018
Location: D
Posts: 154
|
Quote:
Originally Posted by lachinhan
Version 2 seems more confusing and difficult to use than the old version. Because I don't understand it well, I feel it's not as effective as the previous version.
|
Hi Michael,
i feel the same...
The the newer one looks quite difficult for me, too!
In my case i want to align two guitars.
When i open the align takes window , i see a lot of things (parameter) to turn on/off or adjust
Sadly, there are no tutorial on youtube with th e new version.
I would be grateful for help!
|
|
|
01-09-2022, 04:29 AM
|
#79
|
Human being with feelings
Join Date: Oct 2013
Location: Moscow, Russia
Posts: 3,960
|
Quote:
Originally Posted by Finow79
Hi Michael,
i feel the same...
The the newer one looks quite difficult for me, too!
In my case i want to align two guitars.
When i open the align takes window , i see a lot of things (parameter) to turn on/off or adjust
Sadly, there are no tutorial on youtube with th e new version.
I would be grateful for help!
|
Mostly I explained things in this post:
https://forum.cockos.com/showpost.ph...9&postcount=73
I ported mostly old code, so it works pretty same (but with some major improvements which was impossible to fix at the first version). At the same time I didn`t ported settings of the first version (so maybe thats an issue?). I can make some presets for various audio material (or even improve code based on given audio examples).
But unfortunately...I didn`t have any audio examples! So if you send me simple RPP with audio to test, I can setup couple of presets.
If you want to know more tech specifications about every parameter, I can write here.
Last edited by mpl; 01-09-2022 at 04:39 AM.
|
|
|
01-09-2022, 05:00 AM
|
#80
|
Human being with feelings
Join Date: Mar 2018
Location: D
Posts: 154
|
Thanks Michael for the information.
I will read it carefully and make a few attempts.
But where can i find this "window with the circle" to see in here: https://forum.cockos.com/showpost.ph...stcount=73)???
Last edited by Finow79; 01-09-2022 at 05:14 AM.
|
|
|
Thread Tools |
|
Display Modes |
Linear Mode
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -7. The time now is 03:34 AM.
|