How Do You Guys Align a Microphone Signal to a DI one?

interesting…
back some years ago I needed to put the output of my soundcard to the input, record a signal through this round robin and determine the latency of the recorded audio to adjust the time. This is a kind of guessing scenario though…
I have also noticed that when, in cubase, routing audio internally from one track to another for recording purposes, (via a “no output -output” that the recording is not spot on like the first track, hence having me to trim the recording again to fit into place - another guessing scenario
I would like cubase to automatically adjust the latency for these internal doings.
Furthermore, I would like to see an intelligent soundcard handling by cubase, which, allows us to choose our soundcards from a (constantly growing) database list in cubase itself. This would then adjust all the input timings resulting in sample accurate placement of the recordings. This would be a combined effort as all soundcard developers would have to add their latencies to to database. A lot of the soundcard developers already give out this information in the documentations of said devices. (like RME)
To take things further, why not create a AudioTimeCode ( like midi time code, but for audio )

now, back to your question…
by your direct signal - are you referring to the playback in cubase?
and your microphone is going through your fireface?
and, do you want to record your microphone input or for use in a live scenario (in which u can use the direct monitoring of your RME card)
Delay values should be listed in your RMEs manual. (usually approx 1ms out and 1ms in depending on samplerate etc)

hope i didnt confuse:P - just woke up and cant think yet:P
pls let me know how you handle things=)

Right. But that’s not what I mean. I obviously don’t expect Cubase to know the distance I place my mics from each other, or from a Direct Inject signal. However, I would like to know if there’s a way to figure it out after the fact and simply input that distance in the corresponding field (the one found in the track inspector). This would save me the trouble of having to move several recorded clips manually.

If you measure the distance between the 2 you can calculate the delay using the speed of sound. In practice it’s probably easier to record something with a sharp attack so you can easily align the two visually.

The sharp attack is what I always use to help me align these tracks manually. But, like I said, I would like to find a way to calculate these latencies easily and simply input this number in the track (I can’t remember the name of this box in the Inspector. If someone know, please feel free to say it).

I guess I need to study more about the different rulers in Cubase and figure out if there’s a way to input samples instead of milliseconds in this field I can’t remember the name of. That would solve my question.

Thanks!

We’re talking different issues here. In short, I have two Bass tracks (One with the recorded signal of a bass cabinet, using a microphone, and the other is a Direct Inject or D.I. signal). Both tracks capture the same exact take but, as mentioned by Strophoid above, the mic signal will be recorded later than the direct signal due to the distance of the mic from the cabinet. Due to this delay, phase issues arise which become detrimental to the low end (especially in this case since I’m recording Bass), so I need to align these track in order to avoid that and get a punchier Bass sound.

The way I’m achieving this is by recording a high transient Bass sound, such as that from slapping the strings, and using that in order to adjust the latency of the two tracks so that they are phase coherent. That’s basically what it boils down to, and I would like to have a better way to achieve this as mentioned earlier. Hope this clears things up.

ah, i see now… you want a "per channel* record shift in samples.

from the interwebs:
In order to calculate sample length in milliseconds, you’ll first need to know the sample rate of the recording you’re mixing. For example’s sake, we’ll be saying that the recording we’re mixing is at 44.1kHz, which is standard CD-quality digital. You may be using something higher if you’re mixing at 48- or 96kHz.

The formula is quite simple. It is: milliseconds times sample rate.
For example, if your delay between a pair of room microphones and a soundboard feed is 17 milliseconds (representing 17 feet of delay), your formula would be 17 X 44.1.

In this case, you would enter a sample delay of 749.70 samples into the closest source. This would then time-align the sources.

It’s also equally easy to calculate how many milliseconds are in a number of samples. In this case, you’ll use the following formula: samples divided by sample rate.
Using our above example, you would do 749.70 divided by 44.1, which brings us back to the original number, 17.

hope this helps…)

so, find out the delay difference of both tracks in samples (right click ruler and set to sample) - do the maths to convert it to milliseconds and enter that value in your cabinet track. should be a one time deal… if you always mic from the same position…

=)

I already know the math, but thanks :slight_smile:

I guess I was being lazy, hoping to find someone who does this everyday in Cubase and would provide an easier method to calculate the delay without using math. Nothing against manual calculations, but I wanted to find out if perhaps there was an easier way to go about it.

For example, highlight the region that contains the delay, Cubase would immediately tell me what the amount of milliseconds/samples is contained in this region, and simply input this number in the corresponding box. Done! I’m not at the studio right now, but there’s gotta be a way to do this in Cubase. Anyway, I was hoping someone would know this from everyday experience.

so, find out the delay difference of both tracks in samples (right click ruler and set to sample) - do the maths to convert it to milliseconds and enter that value in your cabinet track. should be a one time deal… if you always mic from the same position…

=)

I guess I will have to go with this method otherwise. Thanks for taking the time to help me :slight_smile:

=)

No need to do math. All you really have to do is split the mic track at the very first transient. The length of the empty clip on the left there is your gap.

Or just draw a selection range between the first two transients and the tooltip will probably give you that number directly.

Or, don’t worry about the number, zoom in and just align the mic track to DI track manually by slipping the audio.

Why would anyone ever do manual math for that when the timeline itself is a calculator? :open_mouth: Why delay the track when you can just slip the audio and line it up?

You’re over-thinking it all.

using this way, he wouldnt need to slip or cut audio everytime he records…
It would record correctly everytime…

Exactly! I wanted to take advantage of the delay correction parameter that’s available in the Track Inspector. It would simplify my workflow since I like using Track Versions for different takes which requires me to manually adjust each version the way you have described (by moving the microphone signal forwards in time until it’s aligned with the direct signal). I’m not wanting to overthink it. Just looking to see if there’s a better way to do this, that’s all.

BTW, this reply is directed to Audiocave, even though I quoted ggc above.

And thanks for the Range Tool tip. I’m gonna give that a shot :slight_smile:

Pardon the delay. Been busy working, but I wanted to chime in to say that the Range Tool was exactly what I was missing. Again, thanks for the suggestion Audiocave, and also to ggc for taking the time to help me.

For anyone interested, here’s the easiest way to adjust the delay of your multi-miced tracks:

  • After recording your tracks, set the project ruler to Samples (setting it to Seconds won’t give you accurate results).
  • Zoom In as far as possible and then use the Range Tool to select the area needed to calculate the gap between the mics. The Info Line will give you the exact Range Length (you’ll need to enable the Info Line, if you haven’t already, in order to see this info).
  • Take that number and divide it by the Sample Rate of your project. This will convert Samples to Milliseconds.
  • Now input the result into the Delay box found in the Track Inspector of the corresponding track (in this case, the miced cabinet track). Et voila!


    You could argue that it is much faster to simply move the miced track the necessary amount without having to perform all of these steps. However, as mentioned before, this is not very convenient when you have multiple takes and/or track versions in your project. By taking the time to calculate this delay however, you can enter this number once and everything recorded will be automatically time adjusted. As ggc said, the same offset would apply to ALL similar situations (i.e. Your Bass tracks across different songs in an album) as long as the mic distance is kept the same.

Keep in mind that this procedure is NOT recommended on stereo recordings, where you do want the time differences between the mics to be kept the way they were recorded. It would mess up the stereo effect if you time adjusted them. However, if you had a direct signal as well as a stereo pair of mics for the same source, then you do want to time align the stereo mics to the direct signal the way described in this post. You just need to make sure you treat the stereo mics as one element, and move them together as one.

Well, I hope this helps someone out there who may have the same question.


Take care!

=)
might also want to check out Mautoalign vst:P

Haha! There always a plugin for everything, isn’t there? The only problem with it is that it takes up an insert space, which Cubase doesn’t have a lot of unfortunately.

I think that Steinberg should provide a setting in Cubendo that changes the Delay box so that it can also accept sample values and not just milliseconds. That way you simply enter the number that appears in the Range Length box and be done. That would be great :slight_smile:.

Take that number and divide it by the Sample Rate of your project. This will convert Samples to Milliseconds.

Or just set the ruler to seconds so it will directly give you the milleseconds when you draw the range? :laughing:

Obviously, I already tried that, which is the reason I mentioned that setting the ruler to milliseconds may not give you an accurate number in the first step. Try it, and you will see what I mean. At least it didn’t work in my case.

Take care!

Second post nailed it.

Any calculation would require a precise measurement of this distance. Probably less work to stick to manual alignment.

MAutoAlign is a plug-in which supposedly handles alignment for you, though. (never tried)

However, here’s my Cubase “tip” for refining alignment:

I select the second track and then hover my mouse over that track’s “start” value in the info bar. I use the mouse trackball/scrollwheel to nudge the track sample-by-sample while it’s playing. Using my ears, I find that one sample forward or back can make a pronounced difference in the sound of a doubled (or DI-ed) mic track. Will sometimes spice up an otherwise boring tone!

UA also have a plug-in which does this for you, BTW, called IPB Phase Alignment plug-in for this type of thing, but requires a UAD card, of course. I do have this one, but often just use the mouse wheel method previously described.

I’d check out the MautoAlign plug listed above…


Thanks for the suggestions guys. I just don’t think it’s worth wasting an insert slot per track to have a plugin that does something I can do pretty easy in Cubase (especially with this new method I’ve adopted). The steps I’ve described above take much longer to explain than to perform. It’s pretty fast and accurate, though it would be even faster if Steinberg would also allow the insertion of sample values in the Delay box. That would be very convenient :slight_smile:

BTW, I own a couple of UAD cards (as seen in my signature below). But, like I said, I wouldn’t want to use a plugin on something like this. Even though the VoS preFIX plugin suggested here looks useful, Cubase already does everything that this plugin can do without the need to use a slot. Not only that, but I don’t have to worry about compatibility with Cubase, or even if this plugin is still being developed years down the road. It would suck hard if they stopped developing it for whatever reason. That’s why I only rely on proven companies for my plugin needs (and not many either).

Happy Easter everyone!

I just use AutoAlign, MAutoAlign | MeldaProduction . Much easier. I find it works really well. There is a free trial version available if you’re interested.

+1, Takes five minuttes to learn it, and it’s really useful.