Setting up Audacity for recording practice

[ghmerrill]

After experimenting with some alternatives, I've decided to bite the bullet and set up Audacity (on Windows) to record my practicing at home. Before engaging in hours of research and conversations with my friendly AI, I wonder if someone could just sketch what some critical initial settings should be to get me going with reasonable results.

To start, I plan to use an AT2020 mic -- purely because that's the best mic I have, lying around from years ago when I taught an online university course (certainly not in music). It's also been years since I used Audacity at all, and I've entirely forgotten the details.

The room I'm currently using sucks in various ways -- having a slate floor and abundantly reflective walls (one is mostly glass). Its saving grace (?) may be that it's 20'x20', has a 10 ft. ceiling, and almost half of one wall is open to another larger open space -- but that doesn't seem to help all the sound bouncing off the hard surfaces. I'm not expecting much, but I'd like to get as close as I can to something reasonable in terms of sound veracity of the recording, and then maybe I'll move to a different room.

Any configuration settings for Audacity will be appreciated. My major goal is simply to hear what I "really" sound like. I'm not so much interested in coercing a recording to sound "better", but to make accurate judgments about how I sound and what improvements I may need to make in that. I haven't been at all satisfied with using my Tascam DR-05, although that seemed to work reasonably with tubas in the same circumstances.

[bitbckt]

A dynamic (RE20 or similar) up close might help you tame the room, but that aside…

Start by recording at 24 bit/48kHz. Set your levels around -12dbFS to avoid clipping.

If you’re monitoring live, you might want some simple EQ to clean up the sound a bit. High pass at 80 or 100Hz to start. You’d probably apply a similar HPF at mix time.

That should get you a starting point.

[ghmerrill]

Thanks.

[ghmerrill]

That should get you a starting point.

Okay, that got me a decent starting point. A noticeable improvement on what I was getting previously (even without a high-pass filter, which I haven't gotten around to yet because it apparently requires some extra steps in Audacity).

It also has me now thinking of moving to a Shure SM57 as a potentially better solution (although I'll never know how much better until I try it).

If I go in that direction, it appears that while a simple USB/XLR cable might "work", it would be better to employ some sort of audio interface or an XLR-to-USB "signal adapter". The Shure MVX2U seems pricey for what I want to do (probably a bit more expensive than what I can get the SM57 for!). There's one by Pyle that's about half the price and seems functionally equivalent, and some similar ones around. And of course a whole bunch of small "audio interfaces".

Any thoughts on what would be an acceptably good direction for my limited goals?

[bitbckt]

An SM57 would be entirely appropriate for what you’re trying to accomplish.

There are a variety of single channel USB audio interfaces, though most options open up at two channel (enabling stereo recording). Any one of them will likely work for your needs.

The issue you’re most likely to care about at the low end - and honestly, sometimes the high end too - of the interface market is driver support and stability. You’re on Windows, so you probably want an ASIO driver. RME is the best, and what I use, but likely to be more than you’re looking to spend. I would start by looking at what Focusrite and Universal Audio have that might meet your needs.

[ryant]

If you’re looking for a decent audio interface send me a message. I might be able to help.

rt

[MrHCinDE]

Following this with great interest.

By coincidence I just bought a focusrite 2i2 interface and used Sennheiser MD 421 since I have a pretty bad room, the mic was for sale local to me, and it was a lot cheaper than the alternative of an RE20.

[ghmerrill]

Yeah, I'm looking around for used stuff as well, but haven't found any just yet. Also looking at economy interfaces like the Behringer UMC22 or UMC202HD, and, M-Audio M-Track Solo. The Behringers look interesting, but the 202 seems to be mostly unobtainium at the moment, with projected availability some time in August (if that's even meaningful).

I think I'll kick back and evaluate alternatives for a while, but the info provided here has been very helpful -- in part by increasing my understanding what the issues are.

[AndrewMeronek]

By coincidence I just bought a focusrite 2i2 interface and used Sennheiser MD 421 since I have a pretty bad room, the mic was for sale local to me, and it was a lot cheaper than the alternative of an RE20.

Yeah, getting a simple but decent interface like the Focusrite helps a lot.

[AndrewMeronek]

The room I'm currently using sucks in various ways -- having a slate floor and abundantly reflective walls (one is mostly glass). Its saving grace (?) may be that it's 20'x20', has a 10 ft. ceiling, and almost half of one wall is open to another larger open space -- but that doesn't seem to help all the sound bouncing off the hard surfaces. I'm not expecting much, but I'd like to get as close as I can to something reasonable in terms of sound veracity of the recording, and then maybe I'll move to a different room.

Call me crazy, but with that kind of room it's a pretty simple and noticeable improvement to just store a bunch of random leftover cardboard boxes around the walls. Maybe find a way to hang up a bunch of old towels. It doesn't have to look pretty to sound pretty.

[MrHCinDE]

Might try this, at least in front of my glass door!

[ghmerrill]

Call me crazy, but with that kind of room it's a pretty simple and noticeable improvement to just store a bunch of random leftover cardboard boxes around the walls. Maybe find a way to hang up a bunch of old towels. It doesn't have to look pretty to sound pretty.

I won't call you crazy, but I will call you "unmarried".

[AndrewMeronek]

I won't call you crazy, but I will call you "unmarried".

[MrHCinDE]

Thought I’d share my first effort on the new setup, the main thing I learned is that I need to do a lot of trombone practice before I really start with any of the projects I have in mind.



I’m using a close mic’d Sennheiser MD 421-ii and Focusrite 2i2, and GarageBand on my iPad Pro. What’s really neat is that the interface can be powered by the iPad for a mobile setup, it was an absolute breeze to setup.

If anyone’s interested, the horn is an Olds Opera O-23 with a DE LT 101/D/D8olds.

[ghmerrill]

I need to do a lot of trombone practice

Yeah, I've got mine set up with my currently inferior (but reasonably usable) equipment so it's working well for my purposes. Alas, the results are humiliating, and not because of the inferior equipment. I'm not sure I want to hear it any "better".

[MrHCinDE]

I did think for a while before posting my effort, figured people would turn it off before their ears start to bleed. In the vast majority of cases this forum is very supportive and helpful, constructive criticism is always welcome.

[robcat2075]

The room I'm currently using sucks in various ways -- having a slate floor and abundantly reflective walls (one is mostly glass).

Put a rug on the floor. Get a 10x15 area rug for your 20x20 room. You can always roll it up if it has to not be there all the time.

I'm not expecting much, but I'd like to get as close as I can to something reasonable in terms of sound veracity of the recording, and then maybe I'll move to a different room.

... My major goal is simply to hear what I "really" sound like. I'm not so much interested in coercing a recording to sound "better",

If you already have a decent mic to use, use it. I'm very doubtful about the benefit of a long search just the right mic and A/D converter box combination. A directional mic pointed at your bell is what you sound like. The farther it is from the bell and the more it is not pointed at the bell, the more the room is becoming part of what is captured.

I've been using a basic Samson Q1U USB mic for years to record myself. It captures a faithful rendition of what I sound like.

You don't like the way I sound? Fine. But the mic is capturing a faithful rendition of what I sound like.

That it's a USB mic is not a drawback. It is using the same A?D chips to convert the sound to digital that a separate A/D box would use. The bandwidth of USB far, far exceeds what a mic will need from it.

[ghmerrill]

Put a rug on the floor. Get a 10x15 area rug for your 20x20 room. You can always roll it up if it has to not be there all the time.

As a sonic practical matter, this could be an improvement. As a practical solution, it's a non-starter. There's a lot of traffic through that room, to and from the outside in all kinds of weather. In addition, the idea would be summarily rejected by my wife. Currently, there's a smaller rug (about 6'x8') under my practice location, but I don't think it really affects the sound reflection to any degree.

If you already have a decent mic to use, use it. I'm very doubtful about the benefit of a long search just the right mic and A/D converter box combination. A directional mic pointed at your bell is what you sound like.

I have two mics available to me. One is a good quality AT2020 condenser mic, and the other is smaller unidirectional condenser mic. They really seem to produce the same results, although I need to experiment a bit more with the directional mic, and playing at a closer distance.

The farther it is from the bell and the more it is not pointed at the bell, the more the room is becoming part of what is captured.

True, but this is also a reason why one's choice of a particular type of mic can make a difference.

I've been using a basic Samson Q1U USB mic for years to record myself.

That's a dynamic mic -- which is the direction I'd go in if I decide to get a different mic. So thanks for this suggestion. Apparently the Q1U (the "original" dynamic mic as far as I can see) is now regarded as something of a relic -- which means it (and the later Q2U) are available on Ebay in the $50 range. So this might be worth a try.

But the mic is capturing a faithful rendition of what I sound like.

Well, yes and no. This view rests on certain hidden assumptions about sound, its representation in the recording process and its reproduction for the human listener.

The mic doesn't record anything. There is a stream of transitions from the instrument through the mic through the recording mechanism and through the playback mechanism ... and finally to your ear. The mic performs only one function within that stream. Each step in that stream has an effect on how the sound is finally rendered when you listen to what's been "recorded". What are you using to do the recording and playback? And how do you have that configured? Or are you just recording it with your phone? I haven't tried that, but it would be pretty convenient (for my current purposes) if it worked

That it's a USB mic is not a drawback. It is using the same A?D chips to convert the sound to digital that a separate A/D box would use. The bandwidth of USB far, far exceeds what a mic will need from it.

I don't care too much about that degree of (?) fidelity. But I do regard USB as my preferable mechanism from a practical point of view.

At this point (and with pretty high confidence) that the sound I'm hearing in my recordings isn't a thoroughly accurate rendition of what a human would hear. I'm at least at the point where what I've got is usable to me -- and I've already identified an embouchure problem that I'm now addressing. But I will probably see if I can pick up a Q1U or Q2U just to see what difference there may be.

[ghmerrill]

I've been using a basic Samson Q1U USB mic for years to record myself.

Just as an update ... My Samson Q2U arrived today, and after some experimentation, this seems to make a world of difference. A nice feature of this mic is that it has a 3.5 mm output jack built into it so you can plug headphones directly into that. I'm now pretty confident that I can tell what I really sound like, and the recordings are better.

[MrHCinDE]

This is the last snippet from me on this thread, I promise. Here‘s a first attempt with bass trombone:



I quickly felt somewhat restricted by GarageBand and changed over to Logic Audio thanks to a knowledgable man who pointed me in that direction. I’m still on the iPad with Focusrite 2i2 interface and MD421-ii mic.

[harrisonreed]

If you want an accurate recording of what you sound like, you *must* pull the mic away and put it off axis from the bell. Even better if you use a modest stereo mic technique. There has to be room sound included in the recording.

If you want to do studio style recording, where the sound is really dry and "produced", and you'll mix it with other instruments, then yeah, put up carpets and boxes everywhere. But just know, that isn't what you really sound like. The trombone sound is made up of that initial wave and then also how it flies around whatever room you are in.

There is a sweet spot, with mic placement, where you'll get a clean image of the trombone and just enough of the (likely not ideal) room you are in. That's what you actually sound like. You can cheat and bridge/flesh out the recorded sound with a very modest convolution reverb to improve the room sound for home use (but this won't beat a good room).

Besides the room, which is the biggest piece of the puzzle, mic placement is 98% of the equation, if the goal is to have a realistic sound. Close miking techniques are mostly informed by what people read on popular audio forums, which are all about trying to recreate a studio in the home for recording guitar and vocals. Here you want to eliminate all variables so that you can mix and add effects. But close miking a trombone in a small room with carpeting and sound treatment everywhere is basically placing the audience in that same tiny treated room with you, one foot away, directly in front of the bell. Nobody listens to the trombone that way in person.

Move the mics out -- two meters in front of you, and as high above you as the room allows, up to 3 meters, but leave space between the mic and the ceiling. Try omni first. If the room really sucks or the ceiling is low, try cardiod and add a touch of reverb. The reverb you add should have a shorter tail than the natural sound of the room, to bridge it. That's what you sound like.

[ghmerrill]

If you want an accurate recording of what you sound like, you *must* pull the mic away and put it off axis from the bell.

Done that.

Even better if you use a modest stereo mic technique. There has to be room sound included in the recording.

Haven't tried this. I think I've got enough "connection hardware" to do it. I'm still struggling with some of the UI/UX infelicities and behaviors of Audacity, but getting to the point where it's less irritating and more predictable. So I'll give this a go at some point.

If you want to do studio style recording, where the sound is really dry and "produced", ...

Not really interested in this. I lack a coherent vision of being a Youtube trombone star. I'd just like to be reasonably confident of what I sound like to the octogenarians (well -- those with decent hearing) in our big band New Year's and retirement community gig venues, and to the varied collection of folks in the 300 seat (typically sold out) performances we do at the community arts center. What I'd really like to do is get about a half hour alone on the stage of that venue and record a bit of my playing with the equipment I'm now using. I think I might be able to wangle that at some point. Of course, playing to 300 empty seats won't be quite the same, but it will be another data point.

You can cheat and bridge/flesh out the recorded sound with a very modest convolution reverb to improve the room sound for home use (but this won't beat a good room).

It's difficult to beat physics. Also, my goal isn't to make my recorded sound better, but just to get an accurate idea of what my sound and playing really is in order to correct problems and get better. Again, my experience with trying to do that with the trombone compared to the tuba (in exactly the same situation in the same room) has been a bit surprising -- though I suppose it shouldn't be.

But close miking a trombone in a small room with carpeting and sound treatment ...

This is definitely not my situation -- reasonably sized room, high ceiling, no carpeting (except a relatively small throw rug), and no sound treatment.

Move the mics out -- two meters in front of you, and as high above you as the room allows, up to 3 meters, but leave space between the mic and the ceiling. Try omni first. If the room really sucks or the ceiling is low, try cardiod and add a touch of reverb. The reverb you add should have a shorter tail than the natural sound of the room, to bridge it. That's what you sound like.

Thanks for those pointers. Since I'm approaching a combination of better understanding, and beating Audacity into submission, I can move in the sort of direction you suggest now. Just the change to the Q2U dynamic mic has been significant, and using the mic's own direct headphone output (and comparing that to the recording/playback approach) has given me more confidence that what I'm hearing is what others hear as well -- without any post-recording sound tweaking.

[bitbckt]

Glad to hear you found something that meets your needs, Gary. The most important thing is to get recording and learn what works.

And to an earlier claim, no one is putting $15 ADCs in a product that costs $70 retail. They are not “the same chips”. Whether the result works for you or not is an entirely separate matter.

[robcat2075]

But the mic is capturing a faithful rendition of what I sound like.

Well, yes and no. This view rests on certain hidden assumptions about sound, its representation in the recording process and its reproduction for the human listener.

The mic doesn't record anything. There is a stream of transitions from the instrument through the mic through the recording mechanism and through the playback mechanism ... and finally to your ear. The mic performs only one function within that stream. Each step in that stream has an effect on how the sound is finally rendered when you listen to what's been "recorded". What are you using to do the recording and playback? And how do you have that configured? Or are you just recording it with your phone? I haven't tried that, but it would be pretty convenient (for my current purposes) if it worked

I was so careful to not write "recording" there, yet my labors were for naught.

My view is that mic and mic placement are the biggest variable in what we perceive to come out of the other end.

The actual digital process is pretty much transparent now. Speakers and headphones can add their individual slant, but the commercial ideal is to alter the sound as little as possible and even economical choices do quite well today.

[ghmerrill]

I was so careful to not write "recording" there, yet my labors were for naught.

Yeah ... you almost made it with your use of "capturing" rather than "recording" -- but not quite. Maybe "passing on" would have closer.

[robcat2075]

And to an earlier claim, no one is putting $15 ADCs in a product that costs $70 retail. They are not “the same chips”. Whether the result works for you or not is an entirely separate matter.

Quality analog-to-digital conversion is neither a secret nor expensive process anymore, the significant patents are long-ago expired, and the difference in performance among the various chip options for a particular purpose will be trivial.

[robcat2075]

On Youtube I found a teardown of a Samson CO1U, a mic of similar performance to the Samson Q1U.

The A/D chip inside was an AKM AK5371 whose specs, if accurate , far exceed what any home studio musician is likely to need of it.

One vendor sells it for $4.36 per with an order of 1000 units.

AK5371.jpg

Are the A/D chips in a dedicated XLR interface any better?

[harrisonreed]

On Youtube I found a teardown of a Samson CO1U, a mic of similar performance to the Samson Q1U.

The A/D chip inside was an AKM AK5371 whose specs, if accurate , far exceed what any home studio musician is likely to need of it.

One vendor sells it for $4.36 per with an order of 1000 units.

AK5371.jpg


Are the A/D chips in a dedicated XLR interface any better?

Yes, in a sense. 16 bit 48khz is trivial.

A decent interface will give you 24 or even 32 bit headroom, and 192kHz audio. The interface below used to be cutting edge, but is now just "decent".

https://www.sweetwater.com/store/detail ... lsrc=aw.ds

The big caveat here, of course, is that a human ear can't really get additional benefit from anything beyond 16 bit depth, 44.1kHz bitrate. But for working with audio on a DAW, the bit depth saves far more dynamic data, and going above 44.1kHz bitrate will let you play with timing, etc, without distortion when you mix down to a lower rate. So in that sense, the new converters are way better.

[bitbckt]

As I said, and as Harrison put more concretely, whether a jelly bean ADC works for your needs is a separate matter from the notion that sigma-delta converters are somehow immune to meaningful improvements. Furthermore the circuit around the converter has as much to do with the performance of the end product as the data sheet. Commodity 16/48 may be all you care about, and that’s perfectly fine.

Similar statements could be made about the rest of the mic/preamp/DAC/monitor chain.

[robcat2075]

The Samson USB mics and similarly well-performing mics have been around for 20 years now and yet every time someone here asks about home recording they get directed to overly expensive and overly complicated solutions that will not get them any better result.

A decent interface will give you 24 or even 32 bit headroom, and 192kHz audio. The interface below used to be cutting edge, but is now just "decent".

https://www.sweetwater.com/store/detail ... lsrc=aw.ds

That interface is $1000, They offer to finance it for 48 months. It's like buying a car.

Not a relevant answer to a question such as the OP and many others ask around here... "what do I need to get to make some recordings of myself at home?"

It is not what the OP needs. That $1000 doesn't even get him to a mic.

The Samson mic, for less than $100, is already "decent". More than decent. One would have to search very far to find listeners who can hear a difference... if a difference can even be captured.

The big caveat here, of course, is that a human ear can't really get additional benefit from anything beyond 16 bit depth, 44.1kHz bitrate.

Yup. Rarely mentioned when people ask about what they need to record themselves at home.

But for working with audio on a DAW, the bit depth saves far more dynamic data, and going above 44.1kHz bitrate will let you play with timing, etc, without distortion when you mix down to a lower rate. So in that sense, the new converters are way better.

More stuff that isn't part of the OP's inquiry.

One might as well fret that the USB mics aren't "military grade" or aren't ruggedized for spaceflight.

One will need a super-perfect recording environment and exquisite mics to capture anything that can use more recording definition. No one contemplating recording themselves at home is going to have that. That extra definition isn't something they can use.

[harrisonreed]

But for working with audio on a DAW, the bit depth saves far more dynamic data, and going above 44.1kHz bitrate will let you play with timing, etc, without distortion when you mix down to a lower rate. So in that sense, the new converters are way better.

More stuff that isn't part of the OP's inquiry.

One might as well fret that the USB mics aren't "military grade" or aren't ruggedized for spaceflight.

One will need a super-perfect recording environment and exquisite mics to capture anything that can use more recording definition. No one contemplating recording themselves at home is going to have that. That extra definition isn't something they can use.

Rob, I was responding to your question, not the OP. You specifically asked:

Are the A/D chips in a dedicated XLR interface any better?

To which I specifically gave you a valid response, even addressing the OP's situation (you only likely need 44.1khz if doing zero editing) and possible use cases (wanting to do editing, or adjusting levels of you need to bring up voices in a recorded lesson) where better interfaces would be needed.

I thought that you actually didn't know if the chips in audio interfaces are better than a $4 audio chip. Your question is similar to asking "is a GeForce 4090 really a better graphics card than the integrated graphics on my Chromebook?" Or "Are professional DAWs really any better than Audacity?", or even "Is Maya really any better than Animation Master?". The answer to all of those questions is "yes", not "gotcha! I only play Links 386 on my computer so it's moot!" And these DAWs, like Reaper, and interfaces that can do 32 bit aren't products that are priced prohibitively out of the hobbyist realm (unlike Maya, or Sequoia, which are professional software). Just because something isn't free shareware doesn't mean that you can't use it as a hobbyist.

Also buying a $1000 interface is not like buying a car, at least not in the USA. Sorry, but that is just not true. When is the last time you financed a $1000 car? Or had to register and insure an audio interface?

Anyways, I'm sorry I answered your question. I didn't realize it was rhetorical and that you already knew the answer.