Jun 21, 2016Uncategorized

In this step-by-step tutorial we’ll walk you through the installation process, the creation of your first project, the Automatic Separation and the refinement processes within ADX TRAX 3.0.

For an in-depth look at all the features of the software, please consult the complete User Guide here.
This tutorial has been created using the song Mogador. You can right click here and choose Save Link As to download it and follow along with the steps.

Downloading and installing ADX TRAX 3.0

Login to your ADX account on the Audionamix website where you will see your products and API key to authorize your products. Click Download on your purchased product.

To install ADX TRAX, open the TRAX .dmg file and drag the TRAX icon to the Applications folder.

The first time you open TRAX, you will need to authorize it. The Preferences Window will automatically appear.

In order to authorize your software, you should have your Username and your API Key from your account page, then enter them into the Preferences window.

Creating your first project

Choose to create a new project from the TRAX Project window.

In this example, we will call it “Project 1” and save it into the Desktop.

Once your project is created, import a supported audio file by dragging and dropping onto the interface, or through the File > Import Audio (⌘I) menu command.

In this example, we will import the song “Mogador” that you can download here in order to follow the exact same steps.

Importing an audio file

Once the file is loaded, TRAX will ask you to choose some project-specific options.

Automatic Voice Activity Detection (AVAD) uses an advanced algorithm to detect where a human singing voice is present and extract melodic content only for these sections. With AVAD off, ADX algorithms extract melodic content for the entire duration of the file (this typically works better for instrument melodies rather than singing voices). You can learn more about this dialog here.

For this song we’ll keep the settings at default and click OK.

ADX TRAX Interface

After your initial separation is complete, you can use the keyboard shortcut ⌥A to Zoom All Out and see the entire duration of the file.

Let’s familiarize ourselves with the interface!

At the top of the interface on the Separate Screen, the Vocal and Music tracks display the separated target Vocal waveform, or the separated backing track once a separation is performed. Solo, mute, and volume controls are included.

In ADX TRAX, listening to the Vocal track and the Music track at the same time always equals the Original file. Don’t believe us? Click on the Listen button at any time to compare to the original source file (the full mix).

ADX TRAX Pro Pitchogram

The Pitchogram is a frequency-restricted spectrogram representation of the full audio file, that aims at highlighting the main melody line.

The Pitchogram has been designed in order to easily edit the Pitch Guide Line.

A Pitch Guide Line traces over the exact vocal melody and defines what will be separated. When a file is imported, a Pitch Guide Line is automatically generated by the software (if Automatic Separation or Pitch Guide Preview is checked at import). This Guide Line can be refined. Opacity is controlled by using the Guide Opacity slider.

By moving the Guide Opacity slider to the left, the Pitch Guide Line in blue disappears.Move it to the right to make it re-appear. The Pitch Guide Line in blue represents the melody line that has been automatically identified and extracted. It is possible that the line missed some of the vocal parts you wanted to extract. Let’s see how we can improve the separation from here!

Refining the Pitch Guide Line

During the separation process, the softwares extracts the melody line represented by the Pitch Guide Line. Refining your Pitch Guide is the first step to do in order to improve results.

You can access the Editing tools below the Pitchogram.

To learn more about those tools, and how to use them, you can consult this in-depth page.

Once you’ve edited the Automatic Pitch Guide, click on the Separate button to send your edits to the cloud, yielding a Refined separation.

The ADX Team has created an associated Pitch Guide for the song Mogardor. You can download it from here and import it through the File menu.

New to ADX TRAX 3.0 is the Consonants Marquee tool that indicate where consonants are on a time range on the pitchogram. Consonant annotations are optional, but help identify and extract sounds such as “ch” or “s”, which are often missed when annotating the pitch only. A Consonant Annotation can be placed over any point on a Pitch Guide Line.

Refining the Pitch Guide Line

The Guide Tone<

The Guide is available on the left of the Pitchogram and allows you to listen to the Pitch Guide Line using a generated tone. It is very useful when refining the Pitch Guide. It permits to have direct feedback of your Pitch Guide Line without having to run a separation.

Moving forward: the Process Screen

As soon as your Pitch Guide traces the melody you’re trying to extract as closely as possible, it’s time to move forward to the Process Screen.

You can swap between the Separate and the Process screens from the tabs at the top of the interface, or with the ⌘= shortcut.

The Process Screen allows you to run your separation with different processing options, make a composite track from the results, or apply post-processing options such as the Spatial Isolation Tool.


Using Advanced Processing Algorithms on the Process Screen

To increase the quality of your results, you can access some Advanced Processing Algorithms options by clicking the Processing icon.


In this example file, there is a reverb effect on the voice. In order to maintain the wet / dry balance of the mix, we have chosen to run a separation with “Short Verb” and “Consonants Boost”.

Consonants Boost is a good parameter to combine with Reverb. Indeed, it helps to keep the high frequencies in the vocal file while including the reverb tail.


Comparing Separations

The Process Screen is really useful to compare the different separations you’ve done, and choose the best to move forward with. You can swap between Vocal and Music by using the tabs at the top of the interface (or the ⌘- shortcut), then solo and mute tracks to audition and compare your results. You can also create a composite track by clicking and dragging on the track waveform area while holding the Shift modifier key.

Finishing Your Work

When you are happy with your separation, you can use the Export button to the bottom of the Process screen to bounce the Vocal or Music isolation from your comp track.

You can also access the Bounce functionality from the File menu in order to export your work at any time.

Screen Shot 2016-06-16 at 10.14.28 PM 2

Bounce Mix to Disk will export what you are currently hearing when you hit play (including the Guide Tone on the Separate Screen and multiple soloed tracks on the Process screen). Note that ADX TRAX is non-destructive, meaning that your Music and Vocal files will always sum to the original. This is true EXCEPT when the Bounce Mix to Disk function is used to combine multiple separation tracks from the Process screen.

Bounce Vocal to Disk or Bounce Music to Disk will export differently depending on your current Screen:

On the Separate Screen:

Bounces the Vocal or Music isolation from the track that is currently selected and in view.

On the Process Screen

Bounces the Vocal or Music isolation from your comp track (same as Export button).

Bounce All to Disk combines Bounce Mix, Bounce Vocal and Bounce Music in one command (yielding 3 separate bounces).

Bounce to STEM ADX TRAX 3.0 now allows to export your audio files as STEMS through the File Menu. To learn more about this new format and its possibilities, consult this introduction video.

That’s it!

Have any questions? Ask us at!