Detect audio works but I don't hear the music being inputed

I’m in Ableton and have audio from a track routed to my Scaler track. Interestingly, sometimes Scaler detects the audio and suggests chords, other times not. That’s same track playing and inputing to Scaler. When it does detect the audio it suggests chords, however I can’t hear what’s being played as the source file since it’s going into Scaler. Is there a way to enable that? I can hear the Scaler keyboard notes if I click on them.

Hi @cibby

when used as an instrument, you cannot hear the detected audio through Scaler 2.

However when using ScalerAudio 2 (which is an Audio FX) you can hear the audio you are sending through it.

Is there a good place to see how to use ScalerAudio 2? The videos I saw online showed how to setup audio detection when Scaler is on a track and I can route another track’s audio to it. Specifically I’m not sure how to route audio to ScalerAudio 2 (as an audio FX)

Check out from 7:00"

1 Like

THANK YOU! Exactly what I was looking for :slight_smile:

1 Like

Using Ableton Live 11, there is a similar problem with Scaler 2.5 (Audio) plugin; the audio indicator bar shows strong sound detection, however, detection fails to result a chord. Instead, Scaler 2 remains “waiting for input.” The audio detection works 1 out of 100 times. Help please. Thank you.

You can adjust the threshold of the audio sensitivity with the tiny blue slider located near the detect controls. It makes a difference sliding it to the left or right so experiment. Our resident Scaler analyst @yorkeman has done extensive audio detect testing so maybe he will chime in here.
Screen Shot 2021-12-23 at 9.12.23 AM

I did a lot of testing with the goal of ascertaining how accurate the detection was with specifically orchestrated sequences of chords, and comparing this with other detection applications like Coda etc… However, I did not experience any issues with the detection process itself, bearing in mind the need to perform the operation described by @jamieh above.

With apologies in advance for noting these points, as I am sure they are not relevant

  • the input should be multi-timbral (essentially, you need triadic and above)
  • the file should be flac, wav or aif (if another file types is loaded, it sits waiting for input, and it’s easy to miss a small message that pops up by the ‘suggest’ button setting out the valid types).
  • single notes won’t work
  • legato sequences can product “anomalous” ** results

** possibly unfair, as the chord detection logic depends on changes in input, so legato chords can get missed.

See example of a detection at Scaler audio detection example – Galaxies and beyond …

Thank you. The adjustment you suggested did bring about some improvement. I am able now to get a higher percentage of detection.

I appreciate the info. Thank you.