would love to get feedback/advice on my recordings..
Hi everyone. I have been making some accordion recordings over the past few months. I think they are turning out ok, but I would really like to make them better.

They are all here:  https://www.youtube.com/user/theNewfieFile

I have been using a very simple set up: 2 Audiotechnica 20-20 microphones, a zoom recorder, and a computer with Audacity audio editing software. I set up the mics about a foot away from the right and left hand parts of the accordion, and make a stereo recording. Then I put that into Audacity, and EQ only the left hand to bring out the bass a bit. Then I put a bit of compression and reverb on the whole thing.

I have done some experimenting with mic placement, and different levels of EQ, compression reverb, and panning, to try to get the warmest, most natural sound possible.

So just to set the bar, this would be an example of my idea of a really good accordion recording:


If you compare this to my recordings, this one sounds WAY better. To my ears, mine sound sort of tinny and harsh by comparison.

So some of the specific things I'm looking for feedback on:

Does anyone have any experience with Audacity? Does the reverb suck? No matter how I fiddle with it, it never sounds that great to me.

I struggle with how much low end to add. Are my recordings too bassy and muddy?

When I listen to the 'good' recording, it sounds super heavily compressed to me, especially the left hand. But when I super heavily compress my recordings, they don't sound good. Why?

I realize that I don't exactly have professional gear. I've been thinking about getting some tube pre amps for the microphones. Any thoughts on that? Would it be better to just get better microphones?

I think my room sound might be contributing to the 'tinny harshness' of my recordings. So in my last recording, I hung some blankets up etc to deaden the room. I think it made a tiny difference....

Anyway, that's it. If anyone would like to give my any kind of feedback at all, it would be much appreciated!
I was typing a nice long post out last night but the board crashed... I have some tips I would want to share, so I will redo it later.

What model Zoom recorder do you have and at what quality settings do you record at?

My musical memoires blog/website: http://www.AccordionMemories.com
For what it's worth...:

I have listened to your https://www.youtube.com/watch?v=EnY-xtX-_RQ recording and compared it to your 'ideal' recording. The question is, what do you use for reference / listening back?

First off, the ideal recording is not really my ideal because it has its stereo channels panned hard right and left. Yours is better, although it needs attention because in the 'Tarantella' clip you have both inputs slightly off-centre to the right.

In your case reverb seems overkill. I hear a lot of 'room' and echo in your recording already that could be due to the placement and/or gain of the microphones and material of the room itself. As it is, I wouldn't add more reverb.

I've listened to the oldest track and the new track on my reference monitors and they sound fine to me. In contrast I hear a bit mud in the 'ideal' recording on the bass channel, probably due to EQ. It does not sound clean.

Better pre-amps > better microphones, all depends how much are you willing to spend for this type of recording....whatever you do, make sure you get the most amount of clean gain for your money (the pre-amps should be absolutely noise-less in the volume you are recording).
They sound good to me, Frank! But I don't know anything about this so listen to Jozz, he's the man.
Still waiting on what model of Zoom recorder before posting my opinions, but I do have several articles on recording the accordion on my web/blog site... link in my signature below.

My musical memoires blog/website: http://www.AccordionMemories.com
Hey guys,

I'm using a Zoom H4n Pro.

It is set at 48kHz / 24 bit.  I think on one or 2 of the recording it was set at 44.1kHz / 16 bit.

For reference I am using Sony headphones, the 'can' type that cover your ears. I don't know what model they are, but they $100 headphones. I also have a listen through my home stereo.

Since I have your attention, here is one more sort of technical question I have been wondering about:

After recording, I do things in this order: EQ, compression, reverb, and finally amplify to 0 decibels. Is this best?
that order looks okay, but I would amplify first and mix everything unaltered

there are no set rules (just what sounds good) but generally EQ is done before compression because in EQ there is most of the work and the other way around can lead to changes that would otherwise not be needed
(29-04-2019, 07:28 PM)jozz Wrote: that order looks okay, but I would amplify first and mix everything unaltered

there are no set rules (just what sounds good) but generally EQ is done before compression because in EQ there is most of the work and the other way around can lead to changes that would otherwise not be needed

Actually, did you know that most plug-in modules of most DAWs are designed to run optimally at a certain level?  That level is around -18db to -12db, so normalizing is absolutely always the last thing that should be done... before exporting of course.  Wink

If you talk to most experienced pro studios, they do something similar to what Frank already does, which is  EQ, compression, reverb, and finally normalize, but not necessarily to to 0 decibels, but output to whatever medium they are uploading to... for example:
-24 LUFS for ATSC A85 (Northa American TV)
-23 LUFS for European TV
-14 LUFS for Amazon Echo/Alexa and Spotify
-13 LUFS for YouTube

- SoundClould does not cut down volume (at this time, but all signs show that they will at some point in the future to the -12 to -15 LUFS range), but they recode everything to 128kb MP3 files (a lossy file format) and sometimes when they recode, they will clip and distort your music in the process if you are close to 0db, so output at -7db to -9db for best results there.

Why is all this garbage so important to know?  Well, lets take YouTube for example... if you normalize to 0db, and  after the upload, when it tells you that it is "processing" your video, one of the things that happens is that YouTube "screws" with your audio, recodes it and cuts it down to -13 LUFS.  They do this for one simple reason... so that their commercials are louder than your videos without distortion.  TV stations do it for the same reason... so that we can hear their commercials while we are in the kitchen making a coffee and cut down the movie so that you have trouble hearing it while you are in the same room.   Angry

While your audio is being re-written by YouTube, you can be that the quality drops a nice amount.  So... how *should* people be recording their audio?

1.  You need to be capturing audio at the HIGHEST quality levels your equipment can, all the time if quality is a concern for you.  In the case of the H4N Pro, this is 24-bit 96khz WAV file format in Stereo Mode.  You do this to maintain the highest quality levels throughout your mastering process and finally export based on the destination.  In the case of YouTube video, you want 24bit, 48hz which is basically Blue-Ray quality, a bit above CD quality (which is 44.1khz).  In the case of my Zoom F4, I capture everything at 24-bit 192khz (no typo), master and output at 24-bit, 48khz for YouTube upload.

2.  Even more importantly than the above is... at what levels are you capturing your audio?  If you are coming close to 0db at the time of the recording, you are recording WAY TO HIGH.  Again, most pro studios record at between -18db to -12db, before starting the mastering process.  This is the one main difference between playing live, where you want to maximize audio volume without distortion or feedback.  The goal is the same, however, quality preamps do much better if you record at lower db levels and most importantly, you maximize dynamic range without coming NEAR clipping.  Clipping is the devil... lol

Another reason that my YouTube videos sound so good is that I output at -13 LUFS before uploading to YouTube.  This means two things... YouTube is not recoding my videos to make them quieter and secondly, I am retaining all my quality at 24-bit/48khz soundtracks.  I also take in to consideration the codecs that I use to output cleaner video, but that is a topic for another post.  Smile

First, some of the good news for Frank... the Audio Technica AT2020 mics that you are using are very likely the best budget microphones on the market, with them you can do some pretty amazing things.  They sound GOOD.  Zoom also makes good recorders, so that is never a bad choice, and the H4N Pro is a nice piece of Prosumer level audio recording equipment.

I was listening to your videos (not really videos... more like music with moving notes on the screen... lol), and I did find your recordings a bit lacking in dynamic range and this is definitely not normal AT2020 quality, so I need to find out a little more about your environment, because as you know, the best mics in the world are not going to sound good in a bad environment.  I think that you started to catch on to this with hanging blankets and what not in an attempt to reduce the room's "liveness".

For recording purposes, a near dead room (near zero reverb and a room that doesn't boost either the bass or treble ranges) is the best to start off with.  Like all art, it is best to start off with a blank canvas instead of one with a basic background colour that you cannot change from picture to picture.  If you have a room with a lot of echo, you can affordably make some moveable sound absorption panels (lots of examples on YouTube) to help.  Don't underestimate the value of some carpet too.

Now, how close or how far should the mics be to the accordion?  This is going to depend on (not so surprisingly), the accordion more than most mics.  Basic rule is that you want the mics close enough to capture nothing but the music... but not so close that you can hear the sounds of the keys clacking.  External mics have this as a bit of a small advantage over internal mics.  Whatever clacking sound your accordion has, is going to be picked up by the mics more if they are on top of the mechanics.  Smile  Generally speaking, I have mine 2-3 feet away.  I do that for 2 reasons.. first it sounds good and second, I can make it so that they do not appear in the videos.

You should be hitting no higher than -12db on the meters at the loudest parts of yor music and recording at the highest quality possible.  Then EQ.  

EQ... EQing is both a science and an art.  I really believe that this part can make or break a video, and here, generally speaking... less is more.  It is more common that I lower a range than I raise a range, but sure I do raise things sometimes.  Depending on the tracks I also filter out parts of the bandwidth that doesn't contribute using bandpass filters.

Compression... I am of 2 minds about compression.  Here we are trying to maximize dynamic range, meaning retaining the loudest parts of the song without distortion all while being able to experience maximum lowest volume in the parts where you are playing quietly.  Compression squishes the loudest parts of the song while increasing the volume of the quietest parts of your song... in effect, neatly killing all that lovely dynamic range.  You can quite literally take a very expressive player and make them sound like a computer with overuse of compression as everything sounds the same volume!

I tend to honestly stay away from it, unless for some reason, I did not capture the recording properly (one reason to use the highest quality), or... I am really going in to advanced editing mode and adding my own expression via volume adjustments in the post process.  Professionals do it either way.  Purists or musicians avoid compression, modernists compress the beejeezus out of everything and make their own changes in post.  We are talking recording here... not capturing live shows where compression holds a very important task in helping maximize volume output.

Reverb... I love reverb, I love the ambience, and I like more reverb than what most people do, so to make the music generally more acceptable to a greater range of people, I put up the reverb to where I really like it... and take it down roughtly 30%-40%, just so it is audible but nowhere near overwheming.

Output... again, depends on the destination, but for YouTube, I normalize everything to -13 LUFS and export that, synch it to my video and kill the camera audio track.

For home use, I normalize to -3db just because I never want any part of my stereo chain to come close to experiencing a clipping audio file.  I am very happy with the sound on my Bose speakers, Klipch home audio system with twin 15" subwoofers in an 8.2 surround sound setup and even on my headphones.  I even listen to the video on my cell phone, and generally, within the capabilities of the speakers, it sounds pretty good there too.  When it comes to the files as I want to hear them at home, I EQ and reverb to my own tastes and cut back on things for mass public consumption.

Finally... your video high standard or bar you might want to reach, as mentioned is not all that fantasticly high. When listening to accordions in real life, my ears do not hear right side bass, left side accordionists right hand as I am looking at them. How do they sound like in real life to you? The stereo image in that video is quite distracting and takes away from reality. The sound is pretty good, if a touch too heavy on the EQing and compression and certainly not enough reverb (lol!).

Ok, I am done typing even though I could write a book about this... lol  If interested, I can add (lots) more later.  Smile

One thing that I would want to do, is I would want to get my hands on a raw file of you (Frank), to hear it in it's base form and see what I could make it sound like, that would be a fun project!

My musical memoires blog/website: http://www.AccordionMemories.com
This is all really great information. Thanks so much. I have begun looking through your articles....

The room is a medium sized carpeted living room, with pretty bare walls and minimal furniture. I'm just realizing that this is a big part of what the recordings sound like. I will definitely make a padded 'sound booth' for my next recordings...

My philosophy on mic placement has been: if the mics are too close then some notes right on the right hand pop out too loud, and on the left hand there are changes in volume as the bellows move in and out, so I put the mics just far enough away to make that problem go away. It's usually 1 to 2 feet. I like a bit of key pressing noise, it makes it sound more real to me...I always keep the levels well low, like around -12db.
Your recordings do not sound bad at all. But you have a different accordion than the "reference video" and you use way less reverb as well. Your bass does not sound as deep (EQ can change that) and you have some notes slightly out of tune, like As4 in one direction (the reference video is all with tremolo so we would not hear that there).
The videos have too much channel separation for comfortable earphone listening, but with speakers it should be ok.
Paul De Bra (not Debra...)
Thanks for all that info Jerry, Paul, Jozz and everyone!
(30-04-2019, 11:48 AM)debra Wrote: Your recordings do not sound bad at all. 

Paul did bring up a good point, your recordings are nowhere near horrible, please don't think that I am saying that.  I sometimes get into sharing so much that it sounds like I am being negative, please don't think that at all.  Smile

I have heard (and have personally done) recordings that were a LOT worse.  We all start from zero and over the years as we add to our knowledge and experience, things improve.

Just as an aside, my "studio" is my basement, an "L-shaped "area with carpeting covering 2/3rds of it.  The ceiling is fairly sound absorbent fiberglass ceiling tiles.  My walls are gyprock but most of the time I am shooting in to a cloth backdrop of some kind and that muffles a good amount and furniture and a big computer desk that absorbs and disperses the sound waves around a bit more.

My musical memoires blog/website: http://www.AccordionMemories.com
Hey thanks to everyone who has taken the time to help me out. There is no negativity at all, it's all very helpful.
If my recordings suck, I want to hear about it! haha

You all seem to be in agreement though that my example of a good recording might not be the best one to model myself after. I'm not sure why I think that one sounds so good. I'm curious to hear examples of what you guys think is a 'good' recording.
Tell me what you think of the sound on this video:

While visiting my good friend Ed in Florida this January, I took this video and within a few minutes after he completed playing, had this edit up and running.  This was very much a 3 songs recorded, mastered and edited in a 60 minutes time-frame.  Today I am a little more experienced and the results would be a little better if I had to redo it.

The restrictions I had with this video:
- integrated mics
- mono setup (two mics both on the same channel, no stereo sound)

What I had to work with:
- an amazing musician
- a great sounding accordion with integrated mics
- beautiful location 
- a Sony camcorder and my Zoom F4 audio recorder.

You can compare the sound to the way he was doing it previously (click HERE for an example).

Now, though its still a MONO recording, it sounds very close to how it sounded when I was there listening to him, with the exception or making the bass a touch more punchy and the treble a little more sparkly.  It sounds like an accordion does even if he was there in front of you.

What do you think about this one?

My musical memoires blog/website: http://www.AccordionMemories.com
your recordings don't suck, they're great performances

I think yours is more honest, as is Jerry's. A truer representation if you will, which I like.

but I also think if you want to achieve a certain sound, you must not let anybody steer you in another direction per se

also it would be a learning adventure to dive into the more processed audio, you might just email that guy directly to see what he did for post-processing (I think he even was on this forum a while back)

follow your ideal sound (just don't seperate your stereo as hard as he did... Wink )
(01-05-2019, 07:59 AM)jozz Wrote: ...I also think if you want to achieve a certain sound, you must not let anybody steer you in another direction per se

I think there is a lot of wisdom here, meaning that eventually you are going to create your own style of recording, your own methods and your own sound, but there is nothing wrong with looking at the opinions of others and testing them out for yourself.  Don't follow just one person's advice (unless you REALLY want just that exact "sound"), but use it as part of your learning process.

Like the actually learning process to improve your accordion playing, its not the destination, but the journey along that path that is really the important and fun part.  Smile

My musical memoires blog/website: http://www.AccordionMemories.com
Yes, the o solo mio recording sounds great, very natural sounding.

the reverb on it sounds great!

Forum Jump: