Another plead for Elastic Audio

Community Ideas for Ardour
#1

Hi and nice to meet you all! I’ve just started using Ardour after more than ten years of using Pro Tools, and I’m really loving it :slight_smile: The only thing that is making me doubt about switching is the lack of Elastic Audio or something similar. I know this topic has been discussed before and you’re probably planning on implementing it at some point, but I thought I’d share my experience in case it can add anything to the conversation and hopefully bring this feature a few steps up the priorities list :pray:

I record and produce my songs myself, and I usually lean more towards spending less time practicing my performances before recording, and fixing them later with Elastic Audio, rather than practicing more so later I have to fix less stuff. My reasons for this are that it’s just way faster, and that I don’t apply moral value to skill or how good a performance is. So if the tool is there and it allows me to get the same thing twice as fast, I’ll use it.

When experimenting with Ardour lately, it kind of forced me to practice more before recording, as editing is slower without Elastic Audio. So I thought the less editing I have to do, the better. I did get better takes, but at the end I found myself editing them a lot anyway. No matter how good a performance is, there’ll always be imperfections, and I like to adjust those. I know this is subjective and other producers will prefer to leave those imperfections. But for me (and I presume I’m not alone in this), just knowing that you can make it better is too big of a temptation :rofl:

The main takeaway is that obsessive editing is something that will happen anyway for producers with a workflow like mine, no matter how good the performance is. It’ll just take longer to do without something like Elastic Audio.

Because of this, it’s still more efficient to use Pro Tools for me, and I’ll stick with it for professional projects with a deadline. For personal projects, I’m excited to keep experimenting with Ardour. But if something like Elastic Audio gets implemented, I’m all in!

Thank you so much for developing such a great and customizable open-source software, I’m excited to see what comes next :smiley:

PS: I know there is the region stretching feature, but you have to do a lot of slicing anyway, plus having to change tools constantly, it ends up not being much faster than regular editing for me

#2

Some time ago i had about the same train of thought…but…here’s something i wasn’t even aware of…and it’s pretty old info…like, almost a decade old :). 34:20
https://youtu.be/dk2AMwc4e2k?si=6OduisuTNRkURZ-b
If you lay it all out as @paul did at the event, it raises all kinds of different questions that are simply not easy to answer…

2 Likes
#3

I am curious. If you don’t enjoy practicing or playing your instrument, and also do not like human imperfections, would it not make sense to instead just use AI to generate the sound?

–

As for the actual issue: under the hood elastic Audio is already available, but so far there is no GUI for it (it’s available via Lua e.g. the “swing it” script which moves every other 8th note back). There is one main problem with this: the quality of the stretched audio varies greatly with available free/libre software libraries…

1 Like
#4

I personally have never been a “fix it in the mix” kinda guy. Great performances are essential when initially capturing, making mixing a breeze. I know many people like this however, who would much rather punch in and fix 20 bad notes rather than just play it again.

While there is nothing wrong with this kind of work flow, it seems to be way more time consuming and way less efficient.

3 Likes
#5

Na yeah I’ll never understand the appeal of this kind of musical surgery. In my opinion, it chips away at the humanity in a piece of music and results in something sterile and less animated. But some people just want that I guess…? And maybe I’m just becoming a grumpy old man shouting at clouds with misplaced, out of touch values.

In the context of Ardour, I’d agree. Plus my point about the subtle erasing of the humanity and ‘life’ (for lack of a better adjective) in a given track.

But in a DAW like Reaper, for example, with stretch markers and ReaTune and so on, you can become quite snappy and rapid in the quest for surgical perfection.

I can’t stand the sound of perfectly quantised instrumental parts and über pitch corrected vocals parts myself and using Ardour in place of Reaper made me much more performance-focused and in turn made my music just better. I think…? Thus my gentle advice to the OP would be to embrace the REC screen in Ardour, embrace shifting the focus from hyper surgical micro editing quests for perfection to a more performance-oriented approach and see if that spices things up a little.

Anecdote to close - I thought the lack of CTRL+Z at mixer level in Ardour would seriously frustrate me but it’s actually made me a lot more considerate and methodical when approaching mixing decisions and parameter changes. So you never know - a forced change in approach can be just the ticket to newer, better productions. Maybe.

#6

Yeah, I agree 100%. However, occasionally a “mistake” during a performance can lead to greatness, at least to my ears :slight_smile: That’s what happened to one of my song ideas, as I was jamming a long, I missed a beat and it struck me as the best thing that happened to this song: it turned from a regular 4/4 rhythmic sig to a 15/8 that sounds almost square if you don’t pay attention. This made me reworked the melody and rhythm guitar that became something much more special. The takeway is that whatever you plan to record, play it again and again with openness and free flow. Once you’re in the zone, anything can happen …

2 Likes
#7

[commercial quote] As both an obsessive editor and a live performance focused musician and recorder, I find Ardour’s workflow truly excellent :slight_smile: [/commercial quote]

#8

Having a great monitor mix while tracking is crucial—ideally, it should sound exactly how you want the final result to sound. My best drum takes were the ones recorded with a perfect monitor mix, meaning the engineer didn’t have to make many changes during the final mix. In short: the happier you are with your headphone sound, the better your performance will be.

#9

I’ve been thinking about this for a long time. It’s not realy Ardour’s fault. Even if there wasn’t any technical issues with incorporating the best (ZPlane’s) algorithm, you just can’t incorporate commercial closed code in floss project and retain it’s integrity.
Possible solution would be that some wise third party makes native Linux commercial plugin utilizing ZPlane’s algorithm.
I can survive without this(and detailed manual pitch shifting) feature when it comes to my own music needs. At the same time, this is the first and only reason i’m not recomending Ardour to anyone interested in working with clients. It’s just a mater of time when you’ll get someone who simply cannot reproduce what he imagined and then…you’re in trouble. It’ll all look like only you can’t do what everybody else does with ease.

1 Like
#10

Bounce the stems out to Reaper → do the thing → bounce back to Ardour?

Might need to have your client take a piss and fetch a brew during bouncing, however, lest they get shirty with you for nuking the vibes.

1 Like
#11

Yes, that’s the only okaysh option i can see right now in those cases. But…
It makes you use 2 DAWs, you can’t realy undo-redo if you notice something dosen’t sit right in later stages, so, i can’t recommend it. I’m personly planning of using this exact method if necessary because of my own reasons(i want to use as much floss as possible, and i’ve learned stuff i already know comming to production from underground music scene/DIY mindset, so it’s completely natural to me), but i can’t advise anyone to adopt my way of thinking, especialy in this day and age - doing music production commercialy is realy survival of the fittest. I wouldn’t even recommend that life path to anyone, even without the addition of using floss as the weapon of choice.

To be honest, the only reason i can dedicate some time to learning Linux&Ardour(and KDEnLive) is because for the last 5 years i have a job in marketing (mostly recording/mixing/sound designing audio for tv commercials, and partly doing video editing for some simple tv shows), so, esentialy - lately i can breathe a bit. Life was simply to hectic before that for something like this, and i know a ton of people who just don’t have the luxury of trying anything that isn’t mass used and accepted. I’m aware this probably sounds like to much philosophy for time&pitch correction topic, but i find it to be the core of problems whenever “alternative workflows” are concerned.
We’re trapped :slight_smile: .

2 Likes
#12

Waveform Free is the way to go if you don’t wanna pay for the soft. Pro version (paid) is a really decent DAW, it has complete set of editing tools you will ever need (I think). Even Free version has audio bending (called Time Warp), folder tracks, take management (comping) and many many more.

#13

It’s not about the money (at least not for me). It’s about the essence of it being libre software - i simply love Stallman’s ideas, no kidding. As my old man once said to me - you realy have the problem with the concept of authorities, don’t you?
:slight_smile:

3 Likes
#14

Ah man, and i just realised - that’s about Tracktion Waveform? Maaaan, hell no!..
Worst GUI layout i’ve ever seen :slight_smile: . Tried it about two years ago - deleted half an hour after initial instalation. Waveform&me don’t gel at all :slight_smile: , and i used a lot of different DAWs over the years. Realy started with Fruity&Cool Edit, used Reason for an year, tried Ableton Live (some restricted versions which came with M-Audio soundcards), remastered a couple of tracks using Pro Tools, moved completely to Cubase, did a thing or two in Reaper, also tried Studio Pro, Cakewalk the tried couple of less known daws like SoundBridge etc…even tried ZRythm a month ago (not bad…simple, but not bad at all).
Ah i forgot…i even played with tracker-style software on things like Amiga500 and 286 machines.
Waveform had the weirdest interface of them all. Maybe only that old win thing called “Buzz” was more diffucult to utilise than Waveform :slight_smile: .
Nothing against Tracktion - just my personal opinion. Waveform - nope, not this guy :slight_smile: .

#15

He explains it really well there, it makes sense yeah. Thanks :slight_smile:

1 Like
#16

I love practicing and playing my instruments! I’m a multi-instrumentalist that gets obsessed for weeks or months whenever I get a new instrument :rofl: I play for fun every day!

What I don’t enjoy much, tho, is, once I’ve come up with a cool idea I want to record, spending hours playing the same thing on repeat until I get a level of accuracy that I consider enough for it to be in a finished song (not saying this is the correct mindset, it’s just my personal preference). With Elastic Audio, I can spend more time improvising, composing, coming up with second voices with other instruments, trying different harmonies… the more creative parts of the process, instead of spending all that time practicing the same melody on repeat. It allows me to put more time on the parts of the process where I find more joy.

I still practice to be able to get decent takes of course, but at a certain point the time I have to spend to improve the result juuust a tiny bit increases exponentially, and it makes more sense for me to just record, adjust it with Elastic Audio in 2 min, and jump to the next thing.

I see what you mean with the parallel with AI, but I personally see these as 2 very different things on a core level. Elastic Audio (or “regular” editing, or EQing an instrument to make it sound more pleasant) are tools to work on the stuff you create, while I think AI is about outsourcing everything. Nothing wrong with outsourcing stuff, but with AI the original creators don’t get paid, plus many other ethical problems.

I didn’t know about this Lua script, I’ll definitely check it out, thanks a lot :slight_smile: Someone else shared some context about the “zplane vs open-source time stretching” topic in another comment, and I totally understand it. If this is what we have for now, I’ll work with it! I’m really loving Ardour anyway and I’ll definitely be using it to make some new songs :slight_smile:

1 Like
#17

OK, I understand. I feel the same about Reaper…

1 Like
#18

Also, to be clear I never edit everything 100% to the grid, I like human imperfections and I try to edit my songs in a way that they don’t sound robotic. But I like to be in control of what level of imperfection I have in my songs, so I always edit everything anyway. Even if the take was brilliant and it already sounds good without editing.

I usually adjust it (or leave it if it’s already there) to 60-70% to the grid, but it’ll vary depending on each situation. For example a tin whistle playing in solo over some atmospheric pad will sound cooler with imperfections. But a guitar that I’m recording twice to pan it L and R will sound better if I bring it closer to the grid (still not at 100% tho). Hope it makes sense :slight_smile:

1 Like
#19

Thanks :slight_smile: I actually never adjust stuff 100% to the grid, more like 60-70%, as I don’t like it either when it sounds artificial. What I like is to be in control of that %, which Elastic Audio allows you to do really efficiently (it literally lets you quantize it to a chosen %).

With Ardour I can still do that, it just takes a lot more time because I have to do it via regular editing (I mean slicing and moving regions) rather than automating it. I’m loving Ardour for many reasons, so I’ll roll with it for personal projects where efficiency is not that important, and see what comes up. Maybe I do end up editing less stuff and liking the results :slight_smile: Sometimes having a limitation can give a certain flavour to what you create, and that’s also cool :raised_hands:

2 Likes