Help with audio production (Linux Spotlight) | Proposal

Hi guys, I’ve decided to skip Introduction topic and go straight to the business :smiley: I heard last Big Daddy Linux Live and Linux Spotlights and @BigDaddyLinux aka Roccu :smiley: mentioned that he wants to learn how to boost audio quality of his productions, so I’m here :slight_smile:

Audio production is not my job and I’m far from be professionalism but I know few things that can be helpful and my DAW of choice is Ardour.

I didn’t prepper any sample (they will be unrepresentative anyway because don’t have raw audio files from shows and audio is more or less modified) but for podcast and radio “voice” principles are pretty simple (I’ve heard somewhere that old school audio engineers in radio stations likes to say compress and expand): dynamic compression, pretty aggressive gate and equalizer. Lowpass and Highpass correction will be nice to do as well. I think that Calf Studio Gear will be enough to do the job but if you like you can use other plugins. I’ve found out by accident (daily I record instruments and vocals and don’t combine Ardour with other programs.) some time ago that signal from PulseAudio can be captured separately from microphone which will be helpful with interviews. It’s not difficult it’s only requirement setting recording microphone from audio interface and PulseAudio on second audio track.

Some system tweaks will be required and don’t know how this impact your main machine. We can try to do minimum, I can write script to set daily and recording mode which will revert all changes for audio recording (something similar I’ve made in RecBox) or you can sacrifice one of machines you have behind your back in Linux Spotlight. We will figure out something :slight_smile:

You can use Reaper but this will slow down process because I didn’t use it before and I’ll be not able to provide direct instruction. Only “you need to check somewhere there” or "this option should somewhere there.

Everyone interested in audio production are welcome I’ll try to do my best :slight_smile:

Hi Mr. Glitch

I really appreciate you offering to help. I am very serious about raising the quality of the audio so I am open to any suggestions. I think the proper way to go about this if we do start something like this is to test it out on the test machine for a while to see how well it works. I surely don’t want to mess with the main machine because the last thing I need is something to go wrong and we can’t do BDLL or I miss an interview that I have scheduled.

My test machine is out of action for the moment but should be back up within a week or so. I know Eric Adams has also offered to help with the audio so maybe we could all get on a zoom meeting sometime and work some details out.

Hi Rocco :slight_smile:

Did you hear about AV Linux before? If I remember correctly it can be used in Live session without installing. Depends on how you want to start JACK connection you can use Cadence or configure it in DAW directly:

  • Audio System: JACK
  • Driver: ALSA
  • Device: Your Audio Interface
  • Sample rate: 4800
  • Buffer size: 512 samples
  • Periods: depends on how much machine can handle, I use default

I’m writing this because you can play with without any system tweaking :slight_smile:

If you like, you can also install DAW on your main machine but set ALSA as Audio System. Only for audio editing should work at least acceptable.

If you like, you can also install DAW on your main machine but set ALSA as Audio System. Only for audio editing should work at least acceptable.

I’m worried more about gaming because with tweaking system, audio have priority (probably only heavy), so performance can be not the best. Part of tweaks are for CPU and RAM usage so it’s kinda for plus. Btw, system tweaking can take 10 minutes and it’s not that hard and using JACK is very beneficial. It only looks hard.
Don’t want to through at you too much information at once, so maybe I’ll wait for questions :slight_smile:

If you find time you can do some reading meanwhile. In Tutorials pdf I’ve gathered various articles and YT channels, most is for music but in article you can find how compressor, limiters etc works and if you like to listen dynamic range compression in action you can check out unfa’s tutorials:

If he finds time he can jump in topic. Sure why not :slight_smile:

I prefer to write in forum, Maybe it doesn’t look like but my English is not that good to handle live conversations. Half of the time it will look like “how to say this… yyyy…uuu…hmmm”. Most of the time I help myself with DeepL and LanguageTool plugin :smiley: Time zone also will not be our best friend. If I’m correct you wake up, when I go sleep :slight_smile:

Do you have somewhere raw record of Intro from Linux Spotlight? Length should be enough for my testing, I need to remind myself few things :slight_smile: And maybe worst quality audio track of your guest, 30 seconds will be enough. I’ll try to make some presets with fixed audio cracking, picking, cut echoes etc :slight_smile:

I was a bit bored after drums mixing today and thought to check your audio editing (Shane Kitt of Linux Lads Podcast). And I have some thoughts [ sample here ].
I’m not sure that it is microphones setup or editing fault but your mics are picking. If its mic setup, your mics are set too loud. When I’ve listened podcasts before I didn’t pay much attention.

I’ve mastered it a bit but didn’t fix everything. The proper way is to have both voices on separate track, but I was a bit to lazy to separate Shane on second audio track. Your voice has nice warm low tons and will be worth to enhance (I’m not sure it’s proper word) them in future recordings.

It’s a bit better (I hope, after few hours of working with drum I’m not sure my ears work correctly :smiley: ), but nothing spectacular.

And one more thing, I don’t want to be mean and please don’t take it that way (it’s friendly advice), but you need to work a bit on your diction (00:18).

Regards :slight_smile:

Fine by me. We were planing to get back to this soon anyway. Collaborating would be great since I only know what I know which isn’t all that much!

I’ve seen Audio Adventures with Eric and Rocco few seconds ago (didn’t listen whole record), and what can I say is, you’re talking too much instead of listen audio. What I mean there is too much wondering about how somebody have set EQ’s compressor, which compress Rocco should use it is too much for his usage etc. I will talk about Rocco workflow because I don’t know your workflow Eric.

I know this will be difficult but forget about presets, it’s only takes your focus from building sound/listening and you’re starting analyze how graph is looking.

If my predictions are correct you just pasting tools in queue on mixer strip and setting presets, then mess with EQ and wondering why this not working.

Sound should be built from ground to top which on mixer is top.


     |    |
    | Tool |
     |    |


Let’s say that first will be EQ with enabled Hipass and Lowpass (or Hipass and Lowpass separatlly) filters to cut registers you don’t need in your audio track (It will cut part of background notices and pink noises too).
If you’re set filter and you’re sure that your voice isn’t too flat after filtering go to next step. If you fell that your ears are a bit tired. Make a few minutes break for tea, coffee whatever, when you feel that you start hear normally then back to work.

Next will be gate (noise gate). What this tool does is cutting signal from set value to 0 (“starts” working when you start talking).

Everything from 0 to 1 is muted, then from 1 to 2 signal volume is growing and after 3 it has volume level of microphone. This curves (1 and 3) are it represents how signal will transition. If its arch shape it will transition naturally for example 15,16,17,18.
If it’s sharp transition will be “aggressive”, for example 15,18,22. I don’t know how to describe this better, you will hear. In many cases you don’t need to touch that instead you have specific reason.

Now we can add compressor. I’ve left links above to unfa’s tutorials. I’m sure that he explain it better, so I can skip to EQ.

So, EQ at the end is for shaping sound/voice. You guys have very good microphones so there is no reason to make big adjusts for example MIDs, it shouldn’t be more than one or four decibels up or down. This also don’t mean you need to use every band which is available in EQ. It’s more like an icing on the cake.

If you’re in half-way and you feel that your ears are tired, or they hurt. Stop. Leave it to next day. There is not much sense in that situations to continue because you only bump MIDs down or LOWs higher. And when you back next day it will be mess.

There is no easy way, you need to listen, analyze what is going on and then (after some time) your ears will be tuned to catch all nuances instantly.

Tools in Davinci are pretty cool but in my opinion they are too advanced for newbie/beginner. Install calf plugins and learn in Reaper (if I remember correctly Reaper have own tools, so you can use them also | and they are not that bad if Glen is recommending them, to be precised, equalizer.) for now.
I also suggest to just close doors of your room and practice instead of zoom meetings and use speakers instead of headphones. Sound need space to expand. For example if you start to don’t hear difference just stand up, take one or two steps back and listen. Setting speakers in proper distance is also important.
And from time to time will be good to turn on volume louder than you normally listen playback. Reason is simple, after hour you’ll not necessary will hear everything for example after turning volume up Low registers will be too loud in comparison to MIDs and HIGHs.

Btw, in sample master I’ve used multiband compressor.

I probably forgot about something but writing this took me two hours :smile: Oh, and if you have issue with something and need advice export 30 sec sample (raw and edited), upload on MEGA or other server and leave link here.

1 Like

There is some awesome stuff in this post. Thank you for taking the time to do it, I really appreciate it very much. I wish I had the time to do some work with you on audio quality, I just simply don’t have the time right now.
As far as working on diction from the post above. I take no offense at all, I am not the most precise when it comes to vocabulary and I am sure there are many times where I could word things better but I try to do the best I can.
Thanks for the tips

We can back to topic any time, it’s how forums works :slight_smile: I have email notifications enabled so, even if I don’t hang around here I will know :smiley:
Maybe meanwhile somebody will jump in the topic with some questions or other ideas, so we’ll push topic further with more details and other content creators here will benefit on it by the way.

I’m keeping my fingers crossed for you :slight_smile:

1 Like