[x3d-public] X3Dv4 Audio properties, Sound component and W3C Web Audio: new submission review

Don Brutzman brutzman at nps.edu
Fri Nov 29 10:11:52 PST 2019


0. Attendees: Efi Lakka, Athanasios Malamos, Vince Marchetti (briefly), Don Brutzman.  Regrets: Dick Puk.

-------

1. Much related information and descriptive prior slidesets can be found in the preceding email minutes.

[1] [x3d-public] X3Dv4 Audio and Sound improvements: progress report
     http://web3d.org/pipermail/x3d-public_web3d.org/2019-October/011449.html

...  which in turn contains many references of interest here, especially for W3C Web Audio.

[2] [x3d-public] Fwd: Re: X3Dv4 Audio and Sound improvements: progress report
     https://web3d.org/pipermail/x3d-public_web3d.org/2019-November/011471.html

[3] [x3d-public] Fwd: Re: X3Dv4 Audio and Sound improvements: Efi Lakka presentation 1
     https://web3d.org/pipermail/x3d-public_web3d.org/2019-November/011472.html

[4] [x3d-public] Fwd: Re: X3Dv4 Audio and Sound improvements: Efi Lakka presentations 2, 3
     https://web3d.org/pipermail/x3d-public_web3d.org/2019-November/011473.html

-------

2. We discussed the need for contributors to have signed the Web3D Consortium Intellectual Rights (IPR) agreement, typically as Web3D Consortium members.

	https://www.web3d.org/join

	https://www.web3d.org/sites/default/files/page/Join%20the%20Web3D%20Consortium/Web3D_IPR.pdf

Carefully following these rules ensures that X3Dv4 remains royalty free (RF) open standard, usable for any purpose.

	https://en.wikipedia.org/wiki/Royalty-free

	https://en.wikipedia.org/wiki/Open_standard

Membership has value, for each of us and for everyone (now and in the future).

-------

3. *Design analysis*

Efi showed us her latest work writing up the slideset information.  That document will be posted soon in a follow-up email.

This design for improvements to X3D Sound Component is powerful, quite general and squarely centered atop the W3C Web Audio API Recommendations.  This is very appealing since it means that browser-based implementations have direct access to these advanced capabilities.  We can further expect that implementations will also be available (now or someday) to implementers using standalone applications and other programming languages.

We discussed whether acoustic materials should be integrated with visual materials i.e. within the existing X3DMaterialNode, or provided in a new node. Relevant principle:

	Separation of concerns
	https://en.wikipedia.org/wiki/Separation_of_concerns

There are tradeoffs of course.  For X3D, it is pretty clear that there will be many many more instances of visual Material than acoustic properties.  Example use case:
- A building contains many rooms, some large and some small.
- Many rooms have different acoustic properties, some have identical acoustic properties.
- Each room contains hundreds of different objects with different Material appearance, both directly and via Inline.
- It is impractical (indeed unsustainable) to have repetition of acoustic properties for every single shape.
- Sound properties could be scoped (similar to Lighting global=false or a touch sensor) to affect all peers and children.
- Sound properties within a coordinate reference frame is defined by nearest peer or parent.
- Test case: the inside of a box (or closet) might have different sound properties than the room.  If a user sticks their head inside that volume, then the closer sound properties become active.

This is probably a good test case to develop... might anyone have such an X3D model to contribute for such purposes?  We could animate a viewing path through the building that let a user hear how the sound properties changed appropriately.

Whitepaper definitions of AcousticMaterial look to be an excellent step in that direction.

Efi plans to post an update later today.  It is likely that the next version of this design document will address these concerns in further detail.

-------

4. *Specification*

Getting clear on the design is an important next step.  X3D community participants are requested to review this document and post comments.

Let us please remain focused on the design document so that all concerns are well and thoroughly addressed.  When the document looks pretty well defined, we will extract prose and post (more terse) specification additions in the github X3Dv4 draft for further review.

Dick and I will discuss this approach further at a later time.

-------

5. *Implementations*

Because significant work has been done with X3DOM already, it is reasonable to expect that two implementations (X3DOM and X_ITE) will be possible once the X3D Specification design is stable.

Athanasios and Efi are keen to collaborate with both X3DOM and X_ITE codebase programmers to help implement these advanced capabilities well.

It will be interesting to see how other X3D implementations approach these capabilities - FreeWrl (C++), Castle Game Engine (Object Pascal), et al.

-------

6. *Examples*

We have a number of simple existing examples for VRML97 and X3Dv3 sound model already.  Some might be extended and adapted.

Looking at a variety of locations (X3DOM, X_ITE, W3C Audio, ThreeJS, WebXR etc.) is recommended, since mapping to those gives us further alternatives for comparison.

We should also collect and maintain a handful of reference audio and midi files for use in these test cases.  As ever, open-source licenses throughout.  Many sources are possible, we will curate any reference sounds used for X3Dv4 testing so that they always remain available.  Some simple examples are linked and attached.

More sophisticated X3D model examples are needed to demonstrate, implement and evaluate these capabilities.  Who wants to help?  Speak up, please...

-------

6. *Next steps*

Clearly this work rises to the level of maturity needed for an accepted X3Dv4 contribution.   Efi and Thanos: (shouting) THANK YOU!

We agreed to keep using x3d-public for ongoing dialog, continue meeting approximately every two weeks.  Getting group understanding, comprehension and consensus is really important for this next stage of work.

Our next teleconference is planned for

	Monday December 16: 1100 pacific, 2100 Athens

-------

Very exciting progress, colleagues - thanks one thousand times per second (1KHz).

	https://www.youtube.com/watch?v=PyD9cMarVJk

	https://www.freetranslations.org/voice-translator.html#el for attached "thank you" audio clips.

	Google text-to-speech used for third clip, but they appear to have hidden that link again.

Have fun with X3D Audio!  8)

all the best, Don
-- 
Don Brutzman  Naval Postgraduate School, Code USW/Br       brutzman at nps.edu
Watkins 270,  MOVES Institute, Monterey CA 93943-5000 USA   +1.831.656.2149
X3D graphics, virtual worlds, navy robotics http://faculty.nps.edu/brutzman
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ThankYou.gr.mp3
Type: audio/mpeg
Size: 3971 bytes
Desc: not available
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20191129/bcbea37b/attachment-0003.mp3>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ThankYou.en.mp3
Type: audio/mpeg
Size: 2976 bytes
Desc: not available
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20191129/bcbea37b/attachment-0004.mp3>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: translate_tts_HelloX3D4.mp3
Type: audio/mpeg
Size: 8160 bytes
Desc: not available
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20191129/bcbea37b/attachment-0005.mp3>


More information about the x3d-public mailing list