Audacity, Workflow and Impulse Response WAV's for fingerstyle guitar

[Martin S]

Like a simpleton, I had thought that with Audacity, all you did was:

1.record

2. tweak volumes (in a stereo recording)

3. do a bit of fade in/fade out processing

4. maybe a bit of high pass to get rid of the rumble

5. maybe a bit of editing if there was things I thought I could clean up

6. add a bit of reverb directly to the stereo file

7. maybe a bit of hf eq, to brighten things up as appropriate

8. finally check gains and tweak again if required

9. sit back, listen to the file and think "I need to practice guitar more"

My questions are:

A) It would be good to get other people's takes on their workflows, and see what I'm doing wrong or missing out. What's the optimum procedure so that I get everything in the right order?

B) Turns out that my use of reverb was totally wrong, and I should have created a new copy of the file, added reverb to that, then blended the dry and wet files to get an optimum sound. Now, the normal audacity reverb plugin is criticised as being a bit harsh, so I have downloaded SIR1, which is a free impulse response reverb based plugin, and also downloaded some example impulse response WAV's from the web (they are legion). From a brief foray into this world last night I did think that going down this path will give better results. Question is, does anyone have a favourite impulse response? Anyone got an opinion on response times that work well with fingerstyle?

Generally, I have probably been guilty of adding too much reverb, so I'm going to try and be a bit more sparing in future.

m.

[ukejon]

I'm not sure that adding reverb to a stereo track instead of blending wet/dry is "wrong".

[KevWind]

Might help if you give a brief rundown of what it is you are doing.
What kind and many tracks etc. Just guitar, mono or stereo etc.etc. ? From you work order list it sounds like you are simply doing a stereo guitar track ?

It is not so much something "wrong" as it is some workflow orders may tend to be more efficient .

I am not familiar with Audacity but hopefully you are using "real time" effects. As opposed to having to actually process/render the track to hear the FX

My workflow always involves multiple tracks and instruments
So my workflow tends to be

First before recording I have my custom session/projects templates all ready set up (W/ number of tracks, order of tracks, routing, sends , FX plugins, ,groups etc.) I can then choose what type of template I am going to use and open it up ready to hit record.

1. record

2. edit out any unwanted material

3. adjust basic volume and balance

4.definitely high pass filter directly on the audio tracks

5. apply subtractive EQ directly on tracks

6. apply additive (if any) EQ on tracks

7. apply compression (if needed) on tracks

8. readjust volume and balance (if needed)

9. set up parallel reverb track/s

10. adjust individual audio track sends to send to reverb track and check overall balance

11. decide if I need any global or main output FX

12 Apply any global FX and check for overall balance.


I am uncertain why you would create a new copy of file (unless you are not using real time FX)

[rick-slo]

Audacity does not process effects real time and that is a huge problem IMO. The effects it comes with are not great, to say the least. Also the work flow is relatively clunky and limited compared to most DAWs. That said, it is free which is remarkable considering the time someone(s) put into creating it.

If you are getting into recording and think you will stick with it then look for a more sophisticated DAW and buy it. Personally I use Ableton Live, but there are a number of other choices.

Kev has posted a good work flow (though personally I don't bother with parallel tracks 90% of the time).

[Chipotle]

Effects in Audacity have to be done after the fact, which really forces your workflow in a particular direction. I used it for years but finally jumped to Reaper and am glad I did--I didn't realize how limiting Audacity was making my recording process.

Being able to turn on/turn off effects in real time allows you to hear and fine-tune your changes easily, not to mention the time saved and ability to "undo" things immediately (don't have to copy tracks, process the copy, then blend; or as you mention permanently alter your track, after which there's no going back).

Your steps 2-8... all the post stuff... will become so much easier in Reaper or any other "real" DAW. Audacity is nice for what it is (simple and free) but it sounds like you're ready to move up.

[Martin S]

Ukejon, yes, maybe 'wrong' was the wrong phrase, I realise they are many ways to skin a cat.


Kevwind, I'm just recording solo guitar, using two mics, one channel panned left the other right, so I'm editing a stereo track. I'm not expecting to reach professional standard sounds, but aiming for as good as I can get under the circumstances of not having a soundproofed recording area (and living in a city.) I record single tracks as and when I think I have something to record.
The SIR plug in actually does let you hear the effect of the plug in (although not exactly in real time), and lets you mix dry and wet, without having to make a duplicate track. So it is much better than the standard plugin.
Just to be clear, I was talking about making a duplicate track, rather than a copy of the whole file, I think I'm doing what you call setting up parallel tracks. Thanks for the details of your particular generic workflow.

Rick-slo, a friend of mine talked me into trying a demo version of Cubase (he has a full version) so we could record and swap files easily. Well, that was the idea, but my hardware was just unsuitable, so it was unworkable for me. Watching him set up tracks, process and edit made me quite relieved that Audacity could do enough for my requirements, and in a simple screen. I didn't buy Cubase at the end of the trial, powerful though it is. Horses for courses.

Thanks all for the comments, very constructive.

m.

[Chipotle]

Your workflow isn't wrong, per se. But the reason to copy tracks in Audacity is that it is destructive--if you apply the effect to the original track, you lose the original raw signal. Effects become additive, and you can't go back (e.g., if you EQ and then add reverb, you can't later decide you want to apply the EQ post-reverb; it's already fixed in that order). Even if you do copy tracks, it still can be difficult because you only have one effect at a time and can't change ordering later. A better DAW allows you to do that, and in real time.

[sdelsolray]

Originally Posted by Martin S

Like a simpleton, I had thought that with Audacity, all you did was:

1.record

2. tweak volumes (in a stereo recording)

3. do a bit of fade in/fade out processing

4. maybe a bit of high pass to get rid of the rumble

5. maybe a bit of editing if there was things I thought I could clean up

6. add a bit of reverb directly to the stereo file

7. maybe a bit of hf eq, to brighten things up as appropriate

8. finally check gains and tweak again if required

9. sit back, listen to the file and think "I need to practice guitar more"

My questions are:

A) It would be good to get other people's takes on their workflows, and see what I'm doing wrong or missing out. What's the optimum procedure so that I get everything in the right order?

B) Turns out that my use of reverb was totally wrong, and I should have created a new copy of the file, added reverb to that, then blended the dry and wet files to get an optimum sound. Now, the normal audacity reverb plugin is criticised as being a bit harsh, so I have downloaded SIR1, which is a free impulse response reverb based plugin, and also downloaded some example impulse response WAV's from the web (they are legion). From a brief foray into this world last night I did think that going down this path will give better results. Question is, does anyone have a favourite impulse response? Anyone got an opinion on response times that work well with fingerstyle?

Generally, I have probably been guilty of adding too much reverb, so I'm going to try and be a bit more sparing in future.

m.

As to your workflow, I suggest getting a DAW that allows for realtime use of any reverb tool (plugin, UAD, hardware, algorithmic, convolution) when mixing. That way you can adjust reverb parameters on the fly while you are listening.

[jyaam4]

I found that REAPER has a great work flow. The pricing also is not too bad at all. You could download and try the full version at their website.

[rick-slo]

Originally Posted by jyaam4

I found that REAPER has a great work flow. The pricing also is not too bad at all. You could download and try the full version at their website.

THANKS FOR THE POST!!!!!!!!!!!!!!

[ukejon]

Man, my eyesight just got a whole lot better....

[LSemmens]

Are we a little deaf! Too little love and too much rock and roll!!!!

[Andy Howell]

Here's my workflow — I tend to agree about Audacity (see below).

For Guitar:

I record using stern microphones into a channel in my DAW set to stereo. the gain levels are levelled with the gain inputs on my interface.

EQ — high pass filter and little tweaking, usually a slight cut in the lower mids and, perhaps, a cut in higher mids and boost in highs depending on the piece recorded.

Compression — a threshold to reduce by 5 to 8 db. Ratio, attack and release will be determined by the piece/track. Compression's an important tool but a bit complicated. Graham's Recording Revolution has a great series of free tutorials on compression. Normally I use a modest ratio but find with fast pieces a sometimes benefit from a higher ratio.

Parallel Reverb — the track is sent to an AUX channel where reverb can be applied using faders to dial in the correct amount. this is 'parallel reverb' _ see Doug young's website. there is also parallel compression which works in a similar way but I never use this on acoustic guitar. Depending on the piece I don't always use this but always have it set up on a template.

For Vocal:

Digital Delay — a tiny amount (so you can't here it almost) will fatten up the vocal before we do anything else.

EQ — high pass is usually done at my mic (best done this way I think) and also a PAD if necessary. Subtle tweaking of the view gives more presence — what the mids again.

Compression — again aiming for 5-8 db reduction. Threshold is usually quite modest and attack and release medium as well.

Reverb — sent to another AUX channel for parallel reverb mixing.

Final Mix

Even a simple guitar and vocal track can (but not always) benefit from something additional — subtle fine tuning though.


EQ — fine tuning (usually using a visual graph eq mostly).

Compression — on vocal/guitar mixes can be useful; we are looking for a more modest threshold here — about 3 db.

Reverb _ perhaps something very subtle. The Eventide stereo Room plugin is very interesting here. If your reverb has a pre delay have a play with this.


I have talked about the db amounts in compression. For everything else rely on your ears. Plugin presets are interesting as you can see, for example with ole vocals, where a good starting point might be. However, each track or project will require different settings, some subtle and some more significant.

Sweepable eq bands are very useful here allowing you to make subtle changes — rely on what you hear rather than what you see!

DAW

I tend to agree about Audacity. Using plugins in real time is a must. If you want do a fair amount of recording it will be worth investing in a better DAW. This doesn't have to be a massive expense. If you use Macs Logic is amazing value, comes with a great range of plugins and is difficult to beat on this system.

Most Common Problem

The most common problem I see with people exploring recording for the first time (and indeed still on my stuff) is a tendency to use too much gain at the input stage. Clipping is not nice and usually manifests itself as a 'muddy' sound, especially with vocals. Go easy. When setting recording levels you want 'Pre Fader' metering.

Focusrite Users

Finally, those who have registered their interfaces with Focusrite are currently able to access a series of free plugins from third parties — some of these are very useful. Other manufacturers might do something similar and worth checking.

[Martin S]

Thanks for comments again.

Andy, great tips there. I do prefade monitor (using an Allen and Heath Zed10 as input) , so I know I'm not clipping. I've not used compression (yet). For the way I play, I've been trying to use improving technique to control dynamics, rather than compress, but I don't rule it out in the future.

My conclusion is that I should be looking into getting Reaper...

Hopefully it's possible to use it on a laptop, and doesn't need multiple huge screens like Cubase! :-)

m.

[Andy Howell]

Martin one really great tip I forgot!

Try not to mix right after recording. Give yourself some space.

Then after workington the mix leave it a day. Come back with fresh ears and then make some final adjustments.

At this point leave it. the temptation to twiddle knobs forever is too great.

Come back after a weeks or so and have another look. The great thing about publishing online (Soundcloud an such) is you can replace your file.