Realtime switching between HRTF sets?

[sebasti...@gmx.de]

Dear list,

I am frequently using the SoundScape Renderer to test different sets of HRTFs and to create binaural sound scenes. This works very well, but I wonder if it is possible to switch between several HRTF sets in realtime? We want to compare different HRTFs and find the best matching HRTF for different listeners. Therefore, a direct comparison and switching between HRTF sets would be helpful.

Is this possible with the SoundScape Renderer or do you know a similar tool which allows a comparison and switching of HRTF sets?

Thanks for your help!

Sebastian

[Jens Ahrens]

Hi Sebastian,

There are a few ways to enable switching between HRTF sets. Admittedly, none of them is particularly convenient.

1) Use the Binaural Room Synthesis renderer

This renderer loads a separate set of HRIRs (actually, it's binaural room impulse responses (BRIRs), which are simply non-anechoic HRIRs) via an asdf-file: http://ssr.readthedocs.io/en/latest/renderers.html#binaural-room-synthesis-renderer The original idea behind this is measuring the BRIRs of different source positions inside a given room and then re-sythesizing this room based on the measurement data.

You can assign different HRTFs to different sources and toggle the muting state of the sources to switch between HRTF sets. If I am imagining it correctly, this will allow you to have one single virtual sound source only whose position is locked. But headtracking is possible.

2) Use several SSR instances in parallel

You can run several instances of SSR with the binaural renderer in parallel. Be sure to use the '--name=NAME' command line option to assign different JACK client names to the different instances. I might want to script that if you're using a large number of instances. Also, be sure to disable the IP-interface (option '--no-ip-server') so prevent several instances from trying to listen to the same port. I don't know if different SSR instances can read from the headtracker port simultaneously. Probably not. In other words, there might be headtracking then.

3) Use some other convolution engine

There are few options around. For Max/MSP, there's https://github.com/AlexHarker/AHarker_Externals . For Pd, there is https://github.com/TheTelephone/TheTelephone . However, all of them would require implementing the headtracking yourself.

I'm hoping that helps...

Best,

Jens

--
You received this message because you are subscribed to the Google Groups "SoundScape Renderer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to SoundScapeRenderer+unsub...@googlegroups.com.
To post to this group, send email to SoundScapeRenderer@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/SoundScapeRenderer/29c90314-83fe-44e2-9c86-9e4086ae085f%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[sebasti...@gmx.de]

Jens, thanks for your help. As we need head tracking together with HRTF switching I think we will try to create an own GUI and convolver in Matlab. Since the latest release it allows realtime audio IO, so maybe that enables easy prototyping of a binaural playback engine.

BTW: are there plans to allow 3D playback in the SSR where sources can also be placed at a certain height? This would have been the next thing we wanted to test in our study...

Sebastian

[Jens Ahrens]

Hi Sebastian,

Jens, thanks for your help. As we need head tracking together with HRTF switching I think we will try to create an own GUI and convolver in Matlab. Since the latest release it allows realtime audio IO, so maybe that enables easy prototyping of a binaural playback engine.

Cool!

 

BTW: are there plans to allow 3D playback in the SSR where sources can also be placed at a certain height? This would have been the next thing we wanted to test in our study...

This has been on our todo list for a long time. But as none of the main SSR contributors really needs this functionality, it has a very low priority associated to it. Realistically, it's not going to happen anytime soon unless we find someone who's willing to do it.

Greets,

Jens

 

Sebastian

--
You received this message because you are subscribed to the Google Groups "SoundScape Renderer" group.
To unsubscribe from this group and stop receiving emails from it, send an email to SoundScapeRenderer+unsub...@googlegroups.com.
To post to this group, send email to SoundScapeRenderer@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/SoundScapeRenderer/a52e00e6-2366-4908-9136-65edbbffa544%40googlegroups.com.