XTRAX STEMS: Tips and Tricks

XTRAX STEMS: Tips and Tricks

XTRAX STEMS: Tips and Tricks

So you have just purchased your own copy of XTRAX STEMS, the world’s first automatic stem creator. Now what? Here are few tips and tricks that can be used in XTRAX STEMS.

Choose a song that will be ideal for remixing or sampling. Many users of XTRAX STEMS intend to isolate the vocals of a song and use the newly created a capella stem for remixing. However, if there are no vocals in the selected song you use in XTRAX STEMS, you will still be able to create a drum stem. Because of this, you can also run separations on instrumental songs as long as there are drum elements present. An example of a song that wouldn’t be ideal for remixing would have only piano, bass, and synth because all three instruments would be categorized as the music stem. Import your song by dragging and dropping it into the XTRAX STEMS playlist or use the import command (ctrl+I / command+I).

person playing the drumsTry running multiple extractions simultaneously to save time. XTRAX STEMS allows for up to four different songs to process at the same time. To do this simply open four separate windows of XTRAX STEMS. This will reduce your wait time if you need to separate more than one song.

Compare your results with the different Algorithms. Process your song with the Automatic, Automatic HQ, Generic, and Generic HQ Algorithms and listen to all of your results to determine which is best for your needs. When selecting the Automatic separation option, the vocal separation occurs using an Automatic Voice Activity Detection process. Vocals are separated only when detected. This algorithm is recommended as a good starting point, as it is more likely to discern and separate only the vocals compared to the Generic options. With the Generic algorithm, the Vocals stem will be made of the main melodic content extracted on the entire duration of the file. It can be useful when the automatic voice activity detection misses segments of the voice, or when trying to extract a monophonic and melodic instrument, such as a saxophone solo.  Use the Solo/Mute buttons to listen to each stem or listen to a pair of stems together.

Save your project with ctrl+s / command+S after you run a separation, if you need to return to the project.

Export your stems into an outside Digital Audio Workstation (DAW). Adjustments can be made to the panning and volume parameters inside of XTRAX STEMS, but these will only get you so far. Once the stems are exported (ctrl+E / command + E) and imported into a DAW, you will be able to adjust the stereo separation and add as many effects as desired. Adding a reverb effect to the vocal stem inside of a DAW will help smooth out the vocals to create a more usable stem. You can also edit out instances of unwanted bleed. 

Separate 2 Create! The beauty of XTRAX STEMS is that you can use the separated stems to independently adjust levels or take it to another level and create an entirely new version of the original song. Add your own samples and remix the song and create something unique. Create a mashup by using another song in the same key/tempo. With each element available independently, you can also export in the NI Stems format to mix with the TRAKTOR CONTROL series of controllers. The opportunities are endless with XTRAX STEMS!

 

Tips & Tricks: How to Save the Best Takes Using IDC: Instant Dialogue Cleaner

Tips & Tricks: How to Save the Best Takes Using IDC: Instant Dialogue Cleaner

Tips & Tricks: How to Save the Best Takes Using IDC: Instant Dialogue Cleaner

IDC_BoxL_HiResWe in the post-production audio world know that dialogue – and more specifically, performance – is king. The take where the actress nailed the line, getting just the right amount of authority balanced with feminine charm, is the one the director NEEDS to use. It doesn’t matter that there’s a plane, or bus, or gale force wind in the background. It’s our job as audio engineers to make it work. Sometimes we use an alternate take. Sometimes, the actress reproduces her lines perfectly in ADR. But many times we need to find another way.

Dialogue denoising has traditionally been done with tools that learn a selection of background noise by itself, then use this information to remove this noise from the entire file. This works well for constant, unchanging noise, but not for variable noise like those examples listed above. The quality, frequency, and volume of these noises is always changing, so it’s impossible to learn a piece of noise that will match for the entirety of the clip.

IDC_GUI

But what if we attacked the problem from the opposite direction? That is, what if we could somehow identify the speech instead of the noise, and maintain its quality while suppressing the noise in the background, allowing us to perform speech cleanup quickly and easily? Here’s how to do it with IDC: Instant Dialogue Cleaner.

IDC uses a deep neural network (a type of machine learning) based on thousands of speech files, to identify dialogue in a scene in real-time and separate it from the background content. Simply add IDC as an effect on your track (you can also use it as an AudioSuite plug-in in Pro Tools). Hit play and then turn the background knob to the left to lower the volume. You will hear your dialogue stay remarkably intact, even as challenging background noise is removed!

IDC offers up to 24 dB of noise reduction, and the same amount of control over the level of the speech. You can typically turn the background noise down significantly without affecting the quality of the speech. If you’d like to get more noise out, try increasing the strength parameter.

There are many weapons for dialogue editing and speech cleanup. IDC will make a valuable addition to your arsenal, allowing you to be the hero and save that take!

Watch IDC: Instant Dialogue Cleaner Fix Production Dialogue Issues

Watch IDC: Instant Dialogue Cleaner Fix Production Dialogue Issues

Watch IDC: Instant Dialogue Cleaner Fix Production Dialogue Issues

Audionamix is excited to announce its first real-time product, IDC: Instant Dialogue Cleaner plug-in. Supporting VST and AAX formats, IDC is able to reduce background noise down to -24 db with the turn of a knob. IDC works locally, with minimal processing time, and there is no need for an internet connection. This will be an essential tool for dialogue editors and mixers in reducing a variety of unwanted noise during production recording such as hissing, buzzing, humming, wind, airplanes, refrigerators, and cars. IDC works by using ADX Technology to focus on speech, rather than the traditional approach of other plug-ins that focus on noise. The Algorithms used in ADX Technology are developed using machine learning and a constantly expanding database of speech files.

Through the development of previous Audionamix software such as the TRAX series of products, Audionamix developers wanted to introduce something that would be easy to use for beginners yet powerful enough for professionals. Below are two videos demonstrating the simple and effective noise reducing capability of IDC. A variety of problematic dialogue recordings are used as examples to show the versatility of IDC. IDC will prove to be a valuable plugin for those who need to quickly cleanup dialogue tracks without the headache of processing in an outside program. We are currently testing IDC on a variety of Digital Audio Workstations to ensure maximum compatibility across multiple platforms that offer third party plug-in support. Stay tuned for more announcements regarding IDC!

DEMO 1

Example 1 – Wind noise: Dialogue with heavy wind and no wind filter.

Example 2 – City/traffic noise: Dialogue recorded with cars in background.

Example 3 – Full Soundtrack: Dialogue with water noise in background.

Example 4 – Bus by: Dialogue recorded near a bus stop.

 

DEMO 2

Example 1 – City Noise: Dialogue recorded with a mixture of cars, wind, and pedestrian chatter.

Example 2 – Airplane: Dialogue recorded near the Long Beach Airport.

Example 3 – Buzz: Dialogue recorded with buzzing from a lighting fixture.

Example 4 – Wind: Dialogue recorded without a wind filter.

Example 5 – Hum: Dialogue recorded with a low frequency hum from an electrical appliance.

Venomisto Presents ADX TRAX Pro 3 SP Tutorial Series

Venomisto Presents ADX TRAX Pro 3 SP Tutorial Series

Venomisto Presents ADX TRAX Pro 3 SP Tutorial Series

Hi all,

I’m excited to be working together with Audionamix on some real cutting edge technology. Their source separation solution is the only one I’ve found that really works for creating professional sounding acapellas and instrumentals.

I’ve gone ahead and put together a 5-part video tutorial series on using Audionamix ADX TRAX Pro 3 SP to isolate vocals from a song. In it, we’ll walk through a vocal extraction together, from start-to-finish. Some of the things we’ll cover include:

  • Running the initial audio source separation
  • Refining the pitch guide to achieve a more accurately pitched separation
  • Boosting consonants to extract those hard-to-isolate “S” and “F” sounds
  • Using different algorithms to get the best end result possible
  • Diving into the Spectral Editor to polish your isolation for that last 5–10%

CLICK HERE TO WATCH THE VOCAL EXTRACTION TUTORIAL SERIES

In addition to this tutorial series, I’ve been doing Pro Tip videos. In these, I explain how to use a TRAX power feature. You can find the first 2 I did here, where we cover the Smart Attenuate Tool and the Pan-Specific View.

Finally, I’ll also be live streaming my production process for a new remix track, using TRAX to get the original acapella. I’m pretty excited about it, so stay tuned for more!

Check out the tutorial series today, be sure to subscribe if you enjoy it, and of course, leave a comment if you have any questions or ideas for more videos you’d like to see.

Until next time,

How to Fix Low-level Dialogue in Production Audio with ADX SVC

How to Fix Low-level Dialogue in Production Audio with ADX SVC

Have you ever needed to increase the volume of low-level dialogue in production audio? Audio can be recorded without the proper gain structure, lowering the signal to noise ratio and making dialogue too quiet. When you increase the level of the dialogue, you also increase the level of the noise. ADX SVC (Speech Volume Control) automatically identifies speech and separates it from noisy content, allowing you to control these elements independently.

Here’s how to adjust the speech and noise elements in a mix using ADX SVC. Follow along step-by-step, or watch the video walkthrough:

SVC is compatible with these DAWs. Regardless of your DAW, the same simple steps apply.

STEP 1: Acquire the original mix

In this example, several people are speaking in a scene of a meeting, and the level of the dialogue is too quiet. First load the clip into your DAW – in this example we’re using Cubase. Choose SVC from the track FX list and activate it. Then play through the audio clip. You can also use offline bounce if this is a feature in your DAW.

overview

STEP 2: Set options

Choose the Pitch Range of the speech in your clip. In this case, we select the Male preset from the Speech Settings.

pitch

Then select Separation options to choose which algorithms are applied to the process. For this example we’ll select AVAD (automatic voice activity detection) which will apply the processing only where speech is detected (as opposed to throughout the entire clip). We’ll also turn on High Quality, which will help isolate the voice compared to noise, and HF Boost, which

options3

STEP 3: Separate
Press Separate and the audio is analyzed and the speech identified and separated from the wind noise.

separate

STEP 4: Control

Turn down the background noise and/or turn up the speech to find the appropriate balance between speech and noise.

control

Have an idea for a future video tutorial? Let us know at support@audionamix.com.

How to Adjust Dialogue Within a Full Mix with ADX SVC

How to Adjust Dialogue Within a Full Mix with ADX SVC

Have you ever needed to adjust the level of dialogue relative to music or effects within a film or TV mix, but didn’t have access to the multi-track session? You might need to turn down a backing track, or simply raise the level of the dialogue without affecting the other sounds from the mix. ADX SVC (Speech Volume Control) automatically identifies and separates speech from background content, allowing you to control these elements independently.

Here’s how to adjust the speech and background elements in a mix using ADX SVC. Follow along step-by-step, or watch the video walkthrough:

SVC is compatible with these DAWs. Regardless of your DAW, the same simple steps apply.

STEP 1: Acquire the original mix

In this example, a woman is speaking while an instrumental string piece plays in the background. First load the clip into your DAW – in this example we’re using Reaper. Choose SVC from the track FX list and activate it. Then play through the audio clip. You can also use offline bounce if this is a feature in your DAW.

overview

 

 

STEP 2: Set options

Choose the Pitch Range of the speech in your clip. In this case, we select the Female preset from the Speech Settings.

pitch2

Then select Separation options to choose which algorithms are applied to the process. For this example we’ll only select AVAD (automatic voice activity detection) which will apply the processing only where speech is detected (as opposed to throughout the entire clip).

separationoptions

STEP 3: Separate
Press Separate and the audio is analyzed and the speech identified and separated from the wind noise.

separate

STEP 4: Control

Turn down the background (in this case, the music) and/or turn up the speech to find your desired balance.

control

Have an idea for a future video tutorial? Let us know at support@audionamix.com.