[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [AUDITORY] synthesizing virtual movements of naturalistic sound sources
Dear Valeriy,
If you are concerned with naturally perceived rendering of very near sources, like a bee buzzing around you, loudspeaker sources are not really feasible, unless you are able to go to a WFS system (or NFC-HOA), with a high density of speakers.
Note that very near source positioning (within arm's reach) requires additional processing over and above simple HRTF convolution, unless a near-field HRTF dataset is available.
For very far distances, air attenuation is necessary over and above HRTF. This distance attenuation can be modelled, but should ideally be a function of atmospheric conditions depending how far away the source is.
For detailed distant natural rendering, one may also need to account for general terrain acoustic properties.
For headphone rendering, we have made our research rendered public as Anaglyph (http://anaglyph.dalembert.upmc.fr/), free for all use as a VST plug-in. It is geared very much towards realistic proximity rendering with some basic far distance attenuation.
As with others mentioned, you should be able to automate trajectories using various VST supported hosts (MatLab even supports it now).
Cordially,
-Brian
--
Brian FG Katz, Research Director, CNRS
Groupe Lutheries - Acoustique – Musique
Sorbonne Université, CNRS, UMR 7190, Institut Jean Le Rond ∂'Alembert
http://www.dalembert.upmc.fr/home/katz
-----Original Message-----
From: AUDITORY - Research in Auditory Perception <AUDITORY@xxxxxxxxxxxxxxx> On Behalf Of Giso Grimm
Sent: mercredi 13 octobre 2021 09:44
To: AUDITORY@xxxxxxxxxxxxxxx
Subject: Re: [AUDITORY] synthesizing virtual movements of naturalistic sound sources
Dear Valeriy,
in addition to the suggestion of Lorenzo Picinali you may look at TASCAR
- it is primarily made to simulate arbitrary movements in real-time. It offers rendering methods for loudspeakers and an HRTF simulation.
Examples can be found on our lab youtube channel:
https://www.youtube.com/channel/UCAXZPzxbOJM9CM0IBfgvoNg
Installation instructions (currently Linux only) are on http://tascar.org/
Best,
Giso
On 12.10.21 16:33, Valeriy Shafiro wrote:
> Dear list,
>
> We are looking for ways to generate movement of naturalistic auditory
> sound sources, where we take a recorded sound (e.g. a car or a bee)
> and specify direction, speed and distance/range of its movement.
> Ideally, this would be something publically available and open source,
> but if there is proprietary VR software that can do the job, it would
> be great to know as well . Any suggestions from experts or
> enthusiasts on the list are much appreciated!
>
> Thanks,
>
> Valeriy