AUDIO peak as MIDI trigger

[Max Foxen]

Hello Everyone

I'm looking for a simpel solution. I want to use a microphone set against a door. Wen somebody knocks on the door I want the audio peak to trigger an group in QLAB.

Can you guys help me?

Cheers!

[Matt Padden]

Hi Max,

Look up the recent thread titled ‘MIDI Drum Triggers’ for a way of doing this with a meter plug-in. Alternatively get hold an audio-to-MIDI interface (eg Akai ME35T, Alesis D4 etc). Use these to generate a MIDI event in response to the acoustic event, then assign that MIDI event as the trigger for the cue you want to fire.

Matt

[Matt Padden]

I forgot to add that obviously you’d want to add some arm/disarm cues so that the effect only happens when you actually want it to. Particularly useful for those ‘artillery barrage on a quiet day’ moments...

[micpool]

If you can put up with a bit of latency there may be an OSC method that will allow you to read the level of a mic cue and trigger from it. I dam traveling and don’t have access to a Mac today but will look at it tomorrow.

Mic

[RayZ]

Try using loopback https://rogueamoeba.com/loopback/   (soundflower will also work or jack......)   and a little app that runs in the background on the qlab mac     " beatbars" A2M https://beatbars.com/en/a2m.html...  and, as suggested, add arm disarm cues...

Ray

[micpool]

OK, as promised...

Here's a method that works really well and only requires QLab

I have a Microphone connected to IP 1 of my interface

I want this to meter fully on cue output 1 but I don't want to hear it. You can set the device routing for microphone cues independently to the device routing for audio cues  so I matrix microphone cue output 1 to Octa-capture output 1 at -59dB (effectively off)

(You could also use a cue output that is going to an interface channel that has no amp connected to it.)

There is a timeline group (fire all) with :

a mic cue which switches on the trigger group.

a network cue with a long duration which queries the audio level of the microphone  cue continuously  and uses OSC to  put a number between 0 and 1 in its notes field

a script cue with a repeat loop which continuously reads the number in the notes field  of the network cue and if it is greater than the threshold ( I have used 0.5) starts cue "KNOCK"

I have the microphone resting on my table and if I hit the table the knock effect instantly triggers (with very acceptable latency)

You could stop accidental triggers by putting filters and gates in the effects tab of the microphone cue and set the threshold using the microphone gain on the interface.

Workspace attached. (Be careful: Until you set the device routing matrix for your own output device the microphone will be matrixed at 0 and may  be going at full volume to your speakers)

Mic

[Cait P.]

Thanks for this method! I'm wondering if this can work for multiple cues at the same time? ie if there are a few different mics used to trigger different sounds, what would need to change to have those running at the same time? 

To put this into context, I'm currently designing a show that features skateboarders and an array of ramps etc. and it would be great if each ramp had a contact mic (or some other input device) that could trigger off certain sounds when "activated" by a skater moving over it. 

Any thoughts on how that might be achievable? I've used your method with one plate mic set up so that the threshold only activates when a skater hits the edge of the ramp, however sometimes it doesn't trigger or will trigger before the skater gets there.. Not sure why this is, but I'm going to try with a contact mic instead and hopefully that'll help. I'm not familiar enough with NET/OSC/Scripts etc. to finesse it to suit this specific job. 

TIA if you have any thoughts! 

[jimsta...@zoho.com]

An alternative trigger method could be having a oscillator fed through a security mat at each point you want the skater to trigger the cue.

When the mat is "stepped on" it allows the tone through, becoming your audio trigger. There is no background noise to confuse things.

This from experience 20+ years ago using the side chain of Drawmer noise gates to open vocal microphones when someone stood at the mic.

You'd probably need to build the mats into the ramps, or enjoy watching skateboarders go flying.

Jim

[micpool]

For both these things set up your mic cues as 1 ch starting at x

Re false triggering: you can insert a couple of AUs to EQ and control the dynamics of the mic signal. These could be Apple AUs like bandpass filter and dynamics, These in combination with the gain control of your mic input should be able to sort out your triggering. Bear in mind though that the threshold is purely an audio level. So if the mic receives a louder signal, for instance if the skateboarder jumps 5m from the mic, than when the skateboard glides over the mic, it will trigger on the jump.
If you can’t get consistent audio thresholds for the triggering, then you would have to resort to electromechanical devices like pressure mats, breaking light beams or proximity detectors and comvert those signals to MIDI with external hardware.

Re multiple instances: You can just duplicate the cues and put a number suffix after the q numbers e.g TRIG1, NET1, KNOCK1 etc.

Then change:
The starting ch of the Mic cues so you get a different mic for each grouo
The OSC in the NETx cue to refer to NETx and TRIGx
The script in the threshold cue to refer to NETx and KNOCKx (it would make more sense to rename the knock audio cues to SOUNDx and change the script cue to start these q names)

Good luck!

Mic

On Saturday, March 16, 2019 at 2:34:05 AM UTC, Cait P. wrote:
> Thanks for this method! I'm wondering if this can work for multiple cues at the same time? ie if there are a few different mics used to trigger different sounds, what would need to change to have those running at the same time? 
>
>
> To put this into context, I'm currently designing a show that features skateboarders and an array of ramps etc. and it would be great if each ramp had a contact mic (or some other input device) that could trigger off certain sounds when "activated" by a skater moving over it. 
>
>
> Any thoughts on how that might be achievable? I've used your method with one plate mic set up so that the threshold only activates when a skater hits the edge of the ramp, however sometimes it doesn't trigger or will trigger before the skater gets there.. Not sure why this is, but I'm going to try with a contact mic instead and hopefully that'll help. I'm not familiar enough with NET/OSC/Scripts etc. to finesse it to suit this specific job. 
>
>
> TIA if you have any thoughts! 
>
> On Wednesday, 17 October 2018 08:26:46 UTC+11, micpool wrote:
> OK, as promised...
>
>
> Here's a method that works really well and only requires QLab
>
>
> I have a Microphone connected to IP 1 of my interface
>
>
>
> I want this to meter fully on cue output 1 but I don't want to hear it. You can set the device routing for microphone cues independently to the device routing for audio cues  so I matrix microphone cue output 1 to Octa-capture output 1 at -59dB (effectively off)
>
>
>
>
>
>
>

> (You could also use a cue output that is going to an interface channel that has no amp connected to it.)
>
>
>
>
>
>
>
>
>

[Cait P.]

Thank you both for the considered replies! We may need to resort to something electromechanical, but we'll see :) I'm hoping that by using contact mics it'll avoid louder noises in the vicinity triggering cues, and by placing the contact mics right at the edge of the ramp that the skaters skate over. Wish me luck!

micpool - I duplicated the cues, amended and replaced the suffixes, changed the mic channel, and changed the NET cue to read ..."liveAveragelevel/2 0 1# however the threshold doesn't refresh continuously. I can see that the mic channel is live (the green metering on the input channel) though when looking at the threshold it's a static number.

Is there anything to do with the script or NET cues to direct it to stay open?

I hope this makes sense, sorry it's all a bit new to me!

Thanks in advance,

Caitlin

[Alexander (Mailing List) Taylor]

This is exactly what I've been looking for.  I implemented this for 4 inputs (1 mic cue and 4 pairs of OSC/Script cues), but it's grinding QLab to a halt.  I can run it with two copies of the script cues just fine, but if I add a third the workspace effectively freezes.  Do you have any suggestions on how to remedy that?

Thanks,

Alexander

-- 
Contact support anytime: sup...@figure53.com
Follow Figure 53 on Twitter: https://twitter.com/Figure53
User Group Code of Conduct: https://figure53.com/help/code-of-conduct/
--- 
You received this message because you are subscribed to the Google Groups "QLab" group.
To unsubscribe from this group and stop receiving emails from it, send an email to qlab+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/qlab/3ed54bbf-9b12-4a07-b692-e689521f9393%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

The Right-To-Know Law provides that most e-mail communications to or from School District employees regarding the business of the School District are government records available to the public upon request. Therefore, this e-mail communication may be subject to public disclosure.

[micpool]

How many times per second have you set the OSC to send?

Mic

> qlab+un...@googlegroups.com.

[Alexander (Mailing List) Taylor]

I first tried it at the 30fps of the template, and tried lowering it down to 10fps. Same result, I can only run two at once. 1fps doesn't make a difference either.

Thanks,
Alexander

> To unsubscribe from this group and stop receiving emails from it, send an email to qlab+uns...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/qlab/95f811dd-6965-470a-a6bd-4da8164dd171%40googlegroups.com.

[micpool]

All the mic cues should be 1 ch starting at x, so the OSC query should always be referencing op1

The mic cues will have different cue numbers e.g TRIG1 TRIG2 etc so this is what will change in each OSC address

Mic

[micpool]

Alex and I just went off list to find out what was causing his freeze problem. Basically the workspace didn't like having 4 repeat loops running.  Combining the script for each microphone trigger into 1 script solves the problem. If the script  loop interferes with other things you have going on in a workspace you could probably have a delay before each loop iteration to bring the repeat rate for the script down to around 1/25 sec without any major latency.

Here is the new 4 microphone trigger version:

Workspace attached

Mic

On Monday, March 18, 2019 at 3:38:33 PM UTC, Alexander (Mailing List) Taylor wrote:

To unsubscribe from this group and stop receiving emails from it, send an email to qlab+un...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/qlab/3ed54bbf-9b12-4a07-b692-e689521f9393%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Ritchie Young]

Hey Mic, 

I have just tried implementing this script into a show but I'm having a strange issue I was hoping you could consider. 

The downloaded workspace works fine on my own laptop and the show Mac mini. However, I have added the sound designers show file to the show Mac mini and added the script as a separate cue list on their file. When triggering the group I have noticed the measurement in the notes section of the network cue is static, however the mic level is metering. I have tried copying and pasting the same way into an old show file and it works. 

Any thoughts much appreciated.

Ritchie

[micpool]

This is impossible to diagnose from the info you have supplied in your post.

A screenshot of the cue list pasted into your sound designers show file would probably show what is going on, if you can post that.

Mic

[Ritchie Young]

Hey Mic thanks for responding. 

I ended up just creating a new file and it worked fine. I'm now off that project so no longer have access to the files. 

It was an odd one that I just couldn't figure out. I was assuming maybe something to do with version compatibility or a corrupt file. The script and q labeling was identical but it just didn't function when implemented in the designers original file. 

Like I say thanks for responding though. 

Ritchie