We’ve covered most of the important parts of the Linux audio system, so we’re now ready to use the first major application, a digital audio workstation, or DAW.
There are several DAW programs available for Linux, including Rosegarden, Traverso, and the well-known Audacity editor, but for Jack, the most flexible DAW available was created by Jack’s own designer, Paul Davis. The name of the DAW is Ardour (rhymes with “Arbor”).
Making a New Session
When you start the program, you will see a window asking you if you want to create a new session, or produce an existing session. Unless you’ve already used the program in the past, we’ll create a new session.
This is important: Make sure that the “Create Folder In” option is changed to where you want to have your sessions placed; otherwise, they will all be placed in your home directory, which can get pretty disorganized, depending on how many different sessions you plan to create. For now, let’s create a folder called “Ardour” in your home, and then make the sessions in that folder.
Once the folder is made, then all you need to do is start Ardour, and make a new session. In this case, we’ll call it “Practice” and create its folder in that Ardour folder you just made.
Now, I start with the master bus. This is where all the tracks will be mixed together when being sent to the output, either to the speakers, or to another track strictly meant for the final recording. We want this track to have one or two channels, depending on whether we need stereo or not. If you’re doing voice work, mono is preferable, but if music is your thing, stereo definitely has the benefit. For the sake of practice, we’ll go ahead and select two channels.
The monitor bus is not really necessary, as the master will do fine for the purpose of sending the output to the headphones. No point getting fancy when we don’t need to. Besides, this is one of the features of Ardour I have not yet gotten any practice in using, and since I don’t need it, there’s no reason to have it.
For inputs, I prefer to leave this unchecked. If I have one track, I’ll connect the input myself. If I have multiple tracks, the last thing I want is for the microphone to be connected to every last one. It’s slightly convenient in one use, and badly inconvenient in the rest, so the choice is obvious.
Finally, outputs. No matter what tracks you make, it is extremely convenient to connect them to the master bus, since they can then be mixed as needed. This is especially the case when I’m dealing with 4-channel outputs (as is the case when dealing with ambisonics, a special form of surround-sound encoder plugin); the last thing I want is to have to manually connect new tracks to the master bus, 4 channels each, one channel at a time.
The Editor Window
Okay, so the Session settings are ready, so let’s click “New” at the bottom to create the session.
This is the initial window that comes up with Ardour. There’s a lot to it, but most of it can be ignored for now; we just want to make a basic track recording, no need for a heavy mixed production. However, some of the controls will be very useful in the process of recording.
The first useful set of controls are called the transport. These are the buttons that determine if the project goes, stops, records, rewinds, or fast forwards. All we care about at this point are the tall buttons, ignore everything else.
From the left, you see “skip to beginning,” “skip to end,” “loop-play highlighted range,” “play highlighted range,” “begin playing,” “stop,” and “set record.” Now, the last does not begin recording; in order to record you have to press the “set record” button and then the “play” button, which actually begins the recording process.
The next section is the track view. Of the gray bars, the only ones we need to pay attention to at this time are the “timecode,” “location markers” and the timeline itself. In this case, the timeline we see is for the master bus.
Now, buses and tracks are essentially the same thing in Ardour, with one very big exception: you can’t record on a bus. On a track, though, you will be able to, and when you record, it will show the waveform of the recording in the track bar to the right of the track title and buttons.
The “m” button mutes the track or bus, “s” sets all other tracks to mute, “a” and “g” have to do with automation and track groups, which are beyond the scope of this tutorial. A track will have two more buttons, “p”, the track’s playlist control (allowing you to choose which “take” of a recording you want to listen to), and a red dot that you click to “arm” the track for recording; all tracks with the red button selected will record when the global “set to record” is also set and the “play” is pressed.
The bar under the “master” label is the gain control; this allows you to adjust how much gain is applied to the track. The white line is the 0db mark; at that point, the gain is not increased or decreased, it stays at its current loudness.
The vertical lines between the buttons and the timeline are the volume meters; they will show the volume level of the input. However, as the track is not connected to anything at this time, the bars are completely black.
The “timecode” bar is important when you are attempting to time events; the markings also indicate how close you have the timeline zoomed in. The “Location Markers” bar is essential to a quality recording; if an xrun occurs while recording, an Xrun location marker will be placed at that exact moment, allowing you to check to see if the recording was affected. This bar also marks the beginning and end of a recording session, which affects where the “skip to…” transport buttons jump to.
Above the track, and just under the transport controls, is this group of settings. This determines the mouse mode. The first allows you to resize and move all recordings; great for eliminating mouse clicks at the end of a recording, or moving recording segments together for crossfade (yes, recordings automatically crossfade when overlapped). The second allows you to select only portions of a recording (useful when applying certain filters or adjustments only to parts of a segment). The magnifying glass… well, yeah.
The next item is a very important tool. This allows you to automate the gain on that track. For example, on a music track for a podcast project, you can use this to automatically start decreasing volume when it’s time for the host of the podcast to begin speaking. It also allows you to perform a slow fade at the end of a song. Don’t be surprised if you find you’re using this tool very often.
The next item allows you to stretch or compress a segment without adjusting its pitch. Essentially, this means that the recording goes quicker or slower, and otherwise still sounds like whatever the recording sound was.
Finally, the speaker icon allows you to adjust the playhead as you go, listening and adjusting the speed until the playhead is exactly where you want it to be. the little “v” next to the speaker simply detatches the toolbar; I don’t really find it all that useful.
Now, let me direct your attention to the zoom controls on the bottom-left corner of the window:
As you can see, I have the drop-down selected as “Mouse,” and so should you. This determines where the zoom controls zoom to, which is important; if you want to zoom in on a specific spot in the recording, this will allow you to zoom in where the mouse pointer is. The next two buttons are obvious, and the third will zoom out to where the whole recording region is visible.
Now, we want to make at least one track. Going to “Track” -> “Add Track/Bus” will bring up a window to select the settings for our new tracks.
In this box, you are given the choice of how many tracks you want to add, whether they are tracks or busses, how many channels you want each to have, and whether they are in normal or tape mode.
As mentioned before, the difference between tracks and busses are that tracks can be recorded to. There are advantages to having buses as well, but for now, we want to make a track.
The channel configuration determines how many channels you want your track to have. There are a good number of choices, and they can be changed later. However, for simple voice recordings, mono is recommended.
The track mode should be set to normal. Normal mode allows you to record over a track with a new recording, without losing the contents of the original recording. Tape mode, on the other hand, is destructive, and will destroy what you did on that track; if you make a mistake with the record button, that would become very inconvenient.
Now, you should see a new track appear in the timeline.
We have everything we need to make an initial recording in Ardour, except one thing; we don’t have the microphone connected to the track yet. Let’s fix that.
The first thing we will want to do is go over to the Jack Control application (if you’re not sure what I’m referring to, you might benefit from the tutorial on this site) and bring up the connections window:
As you can see here, everything has been automatically connected, except for the capture device. Before we connect this, I want to focus on the channels for Audio 1.
In this view, you might have noticed that Audio 1 has 1 input port, but on the other side, it has two output ports. This is probably one of the most awesome features of Ardour; a track or bus does not need to have the same number of inputs as outputs; this can allow some extremely flexible audio manipulation with the right plugins. Even without plugins, you can change the balance of the track so it can be positioned in stereo or surround sound! We will cover this shortly, but for now, go ahead and connect the microphone (for me, it’s system:capture_1) to the Audio 1 in port, and return to Ardour.
Now, in Ardour, you might have noticed that the volume meters now show a blue bar, one for Audio 1, and both for master. This is the level of volume that the microphone is picking up. Say something loudly, to make sure the meter doesn’t pass the 3/4 point; you do not want to max out the meter with normal speech, or else you will suffer peaking.
Also referred to as “clipping,” and “distortion,” peaking may work for electric guitars, but not for speech. If it’s too loud, some of the sound will be lost, and your recording will be ruined. If the volume is too much, you may want to adjust your hardware preamp first; this way, there’s no chance of hardware-distorted sound from being recorded.
Before we go to the Mixer, however, one more thing needs to be fixed. At this point, unless you’re using hardware monitoring, you are probably not hearing anything in your monitor speakers or headphones. That’s because Ardour is assuming you’re using hardware monitoring. We can fix that by going to “Options”->”Monitoring”->”Ardour does monitoring”. If you have a high latency in Jack, though, you might want to just keep it set to hardware monitoring, this could be distracting otherwise.
The Mixer Window
While we’re at it, let’s also add some plugins to make sure the sound is clear, but not peaked. In order to do this, we’ll need to use the mixer window. To open the mixer window, select “Window” -> “Show Mixer”.
The mixer window has several components on it. The strips and groups sections on the left are not important; the strips window allows you to hide the mixer strips on the right, while the groups window is handy when you want more than one track or bus to be adjusted at the same time.
You will notice that there is a gray space between the Audio 1 strip and the master strip. This is by design; the window can be resized, but the tracks and user busses will be on the right, while the system busses (master and monitor) will be on the right. This keeps them separate, which is useful when you start working with lots of tracks and busses.
Now, on the track, you can see a lot of different controls. We will focus on them, starting from the top.
The absolute top shows the track’s color (each track will be given a different color to allow recognition on sight; busses are not colored), along with a shrink button on the left and a hide button on the right. The shrink button just changes the button labels to make the strip as thin as possible. Pretty pointless right now, but can be a boon with many tracks. Hide is just that. If you want to show it again, click the strip’s name in the strips bar on the left.
Next comes the track name. Clicking on it will give some options, but I neither know what they do (right now), nor are they needed for a good recording, so we’ll cover them in a later article.
After that is a button with “1”. The name can differ; this is the name of the Jack port the track’s input is connected to. Clicking on this brings up the inputs control, including the port editor. While the port editor is one of the most freakishly-awesome things Ardour can do (you can change the number of available input ports), it’s not needed at this time.
The next item is a button labeled “record.” When it’s pink, it means the track is armed to record when the recording process begins. If it’s not, then you might want to click on it; when we record, we definitely want this track recording.
Next, you will see a black box. This box is the host interface; in here, you can use LADSPA and LV2 plugins, as well as VST plugins, if you’re using the 32-bit version of Ardour (I’m not. Sucks to be me, I guess). Yes, you read this right. VST. As in Steinberg VST. Suddenly, this is looking that much cooler, huh?
You may also notice a second black box at the bottom. This is also a host interface for plugins. The plugins in the top box (pre-processing) will be applied before the audio is recorded, while the plugins in the bottom (post-processing) will be applied when the audio is played back. When it comes out, Ardour 3 will do away with the pre-process host, but for now, we’ll just focus on the post-process host box, in order to ensure original audio does not get modified. I’ll explain later.
Before we get to that, though, let’s make sure the rest is understood. The next 2 buttons, Mute and Solo, were explained previously; one makes this track silent, while the other makes all other tracks silent. The two “-0.0” entries indicate volume levels for the items below them. On the left, you have the gain control; -0.0 means it is neither adding nor removing volume from the track. On the right is the volume meter, and the -0.0 means that as of this point, this track has reached the clipping point at least once, but not crossed it. If it says “-inf”, then it means that it has not processed any sound at all. This number will always be the volume level of the loudest sound this track has processed, regardless of whether it’s actually been recorded or not.
At the bottom of the gain control is a button labeled “M”. This is for automation purposes, and isn’t really important at this point (and also something I haven’t experimented with yet).
At the bottom of the center area are two more buttons, Grp and “Input” Grp allows you to determine which group this track belongs to; when that group is active, adjusting any track in that group will adjust this track, too.
The “Input” button is for the volume meter. The volume meter, by default, will show the volume at the point where the sound data is being input, before any plugins have processed it. By clicking this button, you can change the volume meter from showing the input volume to showing “pre,” the volume of the audio after preprocessing the incoming audio, and “post,” the volume as it’s being sent through the out ports after being post-processed.
You might notice, if messing with the button, that the volume meter splits when “post” is selected. This is because the sound has been transformed into stereo signals at the end of the processing. This is done by the balance control, which is the box after the second plugin box. In this case, there is a little green line in the center. Where the line is determines from which direction the sound will be set to come from when played in stereo. If it’s all the way to the left, then it will come out only the left speaker. Ditto for right.
In the master strip, you will notice that there are two of them. This is because the master bus has two channels going in; one channel is set to be sent to the left speaker (the green bar is all the way to the left), while the other is sent to the right speaker (the green bar is all the way to the right).
When dealing with more than two channels, the two boxes will be replaced by a square representing a surface, on which multiple boxes (speakers) and circles (audio sources) will be placed. I find the whole thing cumbersome, myself; if I need surround sound, there are plugins that do the job more effectively. For a single-source recording, however, there’s no need for them.
Underneath the balance control, there are three buttons, Link, ->, and M. Link is only enabled when you have two-channel balancing active; it ties them together, so that they move as one; when you drag one green bar, the other one moves as well. The arrow is split; either both channels go in the same direction when one is dragged, or they go in opposite directions. M, like the one under the gain control, is for automation of panning, which, once again, isn’t needed for this project.
The following button is like the inputs button, but for outputs instead. You can add or remove channels to the output, and assign them to other inputs. In both cases, you can actually make the connections from the edit option of this button instead of using the Jack Control, but it is easier to see what’s going on from the Jack Control Connections window.
Finally, Comments. This doesn’t actually affect the sound at all; its sole purpose is for you to make notes as to what this particular track is about. If you’re working on a lot of tracks, this can be extremely useful when you need to find a specific sound’s track. It’s gray by default, but when there’s a comment in place, this button will then light up with a different color.
Now that we know how the window is set up, let’s get back to our sound modification. One thing we want to remember: If anything is applied to an original sound before it’s recorded, it’s permanent; there’s no way to undo preprocessing. For this reason, we want to make sure that the original sound is preserved in the recording, so we’ll use the post-process host for plugins, and leave the pre-process box empty.
Right-click on the bottom black box, and select “New Plugin” -> “Plugin Manager.” You should now see a window appear with a list of all the plugins you have installed.
The top box lists all the installed plugins that Ardour found, and shows you their favorite selection, whether they’re hidden, their names, which type of plugin they are, what category they’re assigned to, who made them, and how many inputs or outputs they work with. You can search for specific words by typing in the box, and selecting the appropriate search option from the left drop-down.
Now, the first thing we want is to set up a limiter on the sound. A limiter is an audio processing tool that will, when sound reaches a certain volume level, apply a certain amount of gain, usually negative, to that audio. So, if the threshold is -5, and the volume goes above that, we can apply, say, a -5 gain to the sound, so that it goes back down to -10. This can help prevent clipping by reducing the volume of “loud” sounds, like shouting into the mic.
By typing “Limiter” in the box, you’ll have the full list of plugins with “limiter” in the name. Depending on what you have installed, you might have some in the “Limiter” category, “Limiters” category, or even the “unknown” category.
Now, you’ll need to pay attention to the # of inputs for each plugin; for a mono signal, 2-input plugins will not work, so we are limited to those plugins that use 1 input. Tom Szilagyi’s TAP Scaling Limiter will do fine for our needs, so double-click it, or just click it and then click “add”. It will then show up in the bottom box. Once you have all the plugins selected, click “Insert Plugin(s)” to insert them into the host.
If the name in the host is in parentheses, that means the plugin is being bypassed. Since we need to set the parameters anyways, we’ll just double-click the name.
At the top of all plugin control windows, you will see a preset control; this allows you to save and restore settings to that plugin during a session, and bypass the plugin if you want to hear the sound without the processing. In this case, the limit level and the output volume level will be set to -5.000, so that the limiter can begin the process of preventing clipping on the sound.
Another common tool for audio processing is compression. This sets a threshold for volume, and anything above that volume requires more gain to increase the volume; simply put, compression reduces the difference in volume between the quietest sound and the loudest sound in a recording. This way, you can increase the volume without necessarily passing the clipping point. I have a very soft spot for the Invada compressors, so I will choose “Invada Compressor (mono).”
Definitely a better-looking plugin than the limiter, right? For the most part, I can keep most of the stuff at their default values, but let’s see to it that the compression is applied.
In the plugin, the right-hand meter shows the level of volume; the input along the bottom, and the output along the right. The point where the line bends is referred to as the “threshold” or the “knee.” This is where the compression starts to apply to the volume. The ratio determines how extreme the bend is; the higher the ratio, the more input volume is required for the output volume to increase. If the volume stops flat, then this becomes a limiter; which is pretty pointless if we already have one in place.
So, now we have our plugins in place, we can begin the good part: recording! Aren’t you excited?!
At this point, you can close your plugin windows, and your mixer window; we’re done with those for now. Back at the editor window, we can begin the process of recording.
Once again, make sure that your track is armed to record; the red dot should be in a pink button. You can take a moment to change the track name for now; in larger projects, this helps you keep track of the different tracks’ purposes.
You now want to arm the transport for recording as well, so the top red dot should be in a flashing pink button. The flashing is important; it reminds you that, if you press play, than all armed tracks will record instead of playback. Unless tracks are in “Tape” mode instead of “Normal” mode, this won’t necessarily be a problem (tape mode will cause the existing recording to be deleted), but it can be inconvenient.
Once you’re ready, press “Play” and make your first recording! When you’re done, press “Stop.”
Congratulations! You’ve made your first track. However, we’re not done. After all, what is the point of a track if it is not saved in all its glory. And this specific track is not the one we want to save as a final track; it is, after all, not post-processed, which was the point of the plugins in the first place.
To finish the job, we make another track for the finished product. So, go to “Track”->”Add Track/Bus” and create a new stereo track. Once the track is made, go to the Jack Control Connections window, disconnect the master bus from the speakers, disconnect this track from the master bus, connect the master bus to this track, and then output this track to the speakers (In other words, we want master bus -> this track -> speakers). You also want to disconnect the microphone; otherwise, its input will also be recorded by the final track.
Now, the next thing you want to do is disarm the vocal track and arm the final track. Also, at this point, you want to send the playhead (the red vertical line) back to the beginning; pressing the “skip to beginning” button in the transport controls will do this.
When all this is done, arm the transport record button and press play until you reach the end of the recording.
The final recording is now in place. It’s in stereo, and it has been adjusted using the limiter and compressor. Don’t worry if it looks smaller than the vocal track; the final track is showing two channels in the same space the vocal track is showing one.
To test to make sure the final track sounds good, click on the “s” button and disarm the track. Then skip to beginning and play to hear the recording. Sounds good? Excellent. But… is that a mouse click you hear at the end of the track? That’s not good, it needs to go.
In the case of this track, all you need to do is hold the mouse over the right edge of the colored line at the bottom, and drag until the click has been removed.
Exporting the Final Track
Finally, we need to save the track.
All that you need to do is right-click on the track, select the track’s name (in this case, Final-1.1), and select “Export.” You might notice that Ardour does not export in MP3 format (or maybe it does for you, I’m not sure). If not, I tend to recommend “FLAC” for a good intermediate format; it gets compressed, no audio data is lost, and Audacity can open this file for editing (which can be useful for cleaning noise from the sample). Use the browse button, and give the file a better name than “export.flac”, and then click “Export.”
You are now the proud owner of a new audio file, courtesy of Jack and Ardour.
Now that you have made a simple recording in Ardour, you can start to explore the other features in this program that can take you well beyond what simpler editors like Audacity can manage; per-track plugins and automation in particular can give you a stronger control over the finished product.
Have fun, and make something good!