Board index FlightGear Development Aircraft

Improved Engine Sound work done: P&W JT3D on Boeing 707  Topic is solved

Questions and discussion about creating aircraft. Flight dynamics, 3d models, cockpits, systems, animation, textures.

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby bugman » Fri May 24, 2019 1:42 pm

It's not quite the same as Lego blocks. As far as my knowledge goes - assuming that this is the same as harmonic decomposition - if you use enough higher and higher harmonic modes you can reproduce the original. If we are talking about harmonic analysis and decomposition/transformation here, then I do know what I am talking about because I use it extensively in my day job (which has nothing to do with audio).

I am however uncertain if a harmonic reconstruction of a copyrighted work to extract certain elements of that work (a certain set of harmonic modes) results in a new work in which the original author has no copyright claim. I would have assumed that there would be something written about this legal subject. However I find nothing on the web. So I have no way of knowing if this is legal or not. Well, apart from lots of opinions ;)

Regards,
Edward
bugman
Moderator
 
Posts: 1808
Joined: Thu Mar 19, 2015 10:01 am
Version: next

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby erik » Fri May 24, 2019 2:30 pm

bugman wrote in Fri May 24, 2019 1:42 pm:It's not quite the same as Lego blocks. As far as my knowledge goes - assuming that this is the same as harmonic decomposition - if you use enough higher and higher harmonic modes you can reproduce the original.

The same goes with Lego, just use smaller blocks.
bugman wrote in Fri May 24, 2019 1:42 pm:I am however uncertain if a harmonic reconstruction of a copyrighted work to extract certain elements of that work (a certain set of harmonic modes) results in a new work in which the original author has no copyright claim. I would have assumed that there would be something written about this legal subject. However I find nothing on the web. So I have no way of knowing if this is legal or not. Well, apart from lots of opinions ;)

Well if it's not allowed then at least it is impossible to prove. You get the same re-synthesized output regardless of the sample you used, since you are re-synthesizing the original (because all samples are copies of the same original: the jet-engine in this case).

Erik
Current: Parachutist, Paraglider, Pterosaur, Pilatus PC-9M and variants, ERCO Ercoupe, Fokker Dr.1, Fokker 50, Fokker 100
Less active: Cessna T-37, T-38, Santa Claus. Previous: General Dynamics F-16. Worked on: Wright Flyer
erik
 
Posts: 2245
Joined: Thu Nov 01, 2007 2:41 pm

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby bugman » Sat May 25, 2019 9:55 am

Whether or not it is legal or not is not clear to me. But if I made a lego mickey mouse and used that to sell a product, I'm pretty sure the Walt Disney Company lawyers will come down on me like a ton of bricks! The technique is useful and quite powerful when used by an expert, but I cannot see how this algorithm can be used as a loophole to bypass copyright law. In any case, you simply can get legal permission from the copyright holder.

erik wrote in Fri May 24, 2019 2:30 pm:Well if it's not allowed then at least it is impossible to prove.


Do you realise that you are advocating breaking copyright law with this comment?

You get the same re-synthesized output regardless of the sample you used, since you are re-synthesizing the original (because all samples are copies of the same original: the jet-engine in this case).


If the technique is what I think it is, then the harmonic modes you extract will be highly dependent on the quality of the recording device. Artefacts in the recording will leave a 'fingerprint' in the reconstructed sound that experts in the field would be able to identify if they possess the original sound. This is a mathematical processing that is highly dependent on what you feed into it, and the results from different sound sources might sound similar but they will have subtle identifiable differences.

Regards,
Edward
bugman
Moderator
 
Posts: 1808
Joined: Thu Mar 19, 2015 10:01 am
Version: next

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby Thorsten » Sat May 25, 2019 10:09 am

I have a rather basic question:

Is there a Fourier analysis done on an actual recording of a sound file with unclear copyright and then a synthesis based on that decomposition?

Or is there a synthesis being done based on a description of Fourier coefficients published somewhere?

I'd assume that since the first is explicitly based on a copyrighted file, it can't easily be made someone else's copyright (I take a photograph of a copyrighted work of art - doesn't give me copyright. I run a blur filter over the photograph - still not - only if I use small parts which combine to a different whole, say a collage, I can claim fair use for the parts and copyright as a new artwork for the whole - at least that's my understanding) .

Whereas a synthesis from scratch based on technical information is arguable just using real facts for a new creation (I can legally make a 3d model of ISS based on available size data).
Thorsten
 
Posts: 12490
Joined: Mon Nov 02, 2009 9:33 am

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby erik » Sat May 25, 2019 1:19 pm

bugman wrote in Sat May 25, 2019 9:55 am:
erik wrote in Fri May 24, 2019 2:30 pm:Well if it's not allowed then at least it is impossible to prove.

Do you realise that you are advocating breaking copyright law with this comment

I'm not advocating breaking the copyright since what it does is not a matter of recreating the original file but rather recreating the colors and manipulate them to match your needs.
The re-synthesized can do the following which the original could not:
    1. loop seamlessly without any problems.
    2. keep a straight pitch forever
    3. keep a constant volume forever
    4. manipulate the above two features to match engine characteristics under various conditions with just one sample file

If you know music it's all about recreating the timbre of the jet engine and manipulate it using filters and effects. This timbre file could be just milliseconds long.

If you want to be particularly safe you could recreate the timbre using not just adding sine-waves but also exploit the characteristics of the sawtooth, impulse and square wave to match that of the analyzed signal. Exactly matching the timbre is seldomly necessary. You can get close enough to be convincing.

So no, I'm not advocating breaking the copyright law. It just might look that way for someone who doesn't know the matter.
Current: Parachutist, Paraglider, Pterosaur, Pilatus PC-9M and variants, ERCO Ercoupe, Fokker Dr.1, Fokker 50, Fokker 100
Less active: Cessna T-37, T-38, Santa Claus. Previous: General Dynamics F-16. Worked on: Wright Flyer
erik
 
Posts: 2245
Joined: Thu Nov 01, 2007 2:41 pm

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby erik » Sat May 25, 2019 2:07 pm

Let's try to explain it in a different way.

What you want to do is to deconstruct the engine in it's various components: The high whine from the core, the air-sucking sound from the intake (which is caused by a bit of phasing), the thunder noise from the exhaust and recombine them based on the various settings of the engine: core speed, throttle setting.

The end result has nothing in common with the sampled engine sound, but everything with the characteristics of the engine itself.

If you think this is copyright infringement, try suing google (and others) over analyzing the web to train AI.

Erik
Current: Parachutist, Paraglider, Pterosaur, Pilatus PC-9M and variants, ERCO Ercoupe, Fokker Dr.1, Fokker 50, Fokker 100
Less active: Cessna T-37, T-38, Santa Claus. Previous: General Dynamics F-16. Worked on: Wright Flyer
erik
 
Posts: 2245
Joined: Thu Nov 01, 2007 2:41 pm

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby erik » Sat May 25, 2019 2:22 pm

I wish I had a way to get AeonWave working and tested on iOS, the n this could all be done use XML files:

https://github.com/adalinbv/aax/blob/ma ... ntake.aaxs
https://github.com/adalinbv/aax/blob/ma ... haust.aaxs
https://github.com/adalinbv/aax/blob/ma ... e-fan.aaxs

Erik
Current: Parachutist, Paraglider, Pterosaur, Pilatus PC-9M and variants, ERCO Ercoupe, Fokker Dr.1, Fokker 50, Fokker 100
Less active: Cessna T-37, T-38, Santa Claus. Previous: General Dynamics F-16. Worked on: Wright Flyer
erik
 
Posts: 2245
Joined: Thu Nov 01, 2007 2:41 pm

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby Johan G » Sat May 25, 2019 10:18 pm

erik wrote in Sat May 25, 2019 1:19 pm:[...] what it does is not a matter of recreating the original file but rather recreating the colors and manipulate them to match your needs.

erik wrote in Sat May 25, 2019 2:07 pm:What you want to do is to deconstruct the engine in it's various components: The high whine from the core, the air-sucking sound from the intake (which is caused by a bit of phasing), the thunder noise from the exhaust and recombine them based on the various settings of the engine: core speed, throttle setting.

Much like using reference photos while making a livery from scratch in Inkscape or GIMP, rather than "painting" the livery by stretching photos from the web over a paintkit. :wink:
erik wrote in Sat May 25, 2019 2:07 pm:The end result has nothing in common with the sampled engine sound, but everything with the characteristics of the engine itself.
Low-level flying — It's all fun and games till someone looses an engine. (Paraphrased from a YouTube video)
Improving the Dassault Mirage F1 (Wiki, Forum, GitLab. Work in slow progress)
Some YouTube videos
Johan G
Moderator
 
Posts: 6629
Joined: Fri Aug 06, 2010 6:33 pm
Location: Sweden
Callsign: SE-JG
IRC name: Johan_G
Version: 2020.3.4
OS: Windows 10, 64 bit

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby bugman » Mon May 27, 2019 9:17 am

Like I said, I have no idea if this is a copyright issue or not. My opinion is that it probably is not. However I would like to be sure. My concern is because I took over as one of the maintainers of the FGAddon repository from Clement de l'Hamaide (though that job is now done by multiple people) and, if there is a lawsuit, I'll be in the cross-hairs of the lawyers. I don't want to be extradited to the US, India, Brazil, China, or anywhere else for that matter.

I would be interested hearing from ghedger42 about the methods and techniques used, the software used, etc. Especially for the wind sounds. There are some things that confuse me, for example:

ghedger42 wrote in Mon May 13, 2019 8:04 pm:...which makes for much smaller loops that are clean-room engineered from scratch (based on harmonic analysis and Fourier recombination).


Is this clean-room design or Cleanroom software engineering? I assume clean-room design, as that is the technique used to defend against a copyright infringement proceeding (a lawsuit). But as a copyrighted sound was used as a starting point, is this really a clean-room design? If so, how? I would like to hear from ghedger42 about this.

I have been performing a lot of web searches for a legal opinion about this. Writing such articles is what law firms do to distinguish themselves. I would have assumed that this signal processing technique would be extensively discussed. There are plenty of high quality, peer reviewed journals that cover such subject matter. But, I assume due to a lack of technical details, I cannot find any literature yet. There should also be some case law on this subject.

Regards,
Edward
bugman
Moderator
 
Posts: 1808
Joined: Thu Mar 19, 2015 10:01 am
Version: next

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby radi » Mon May 27, 2019 10:07 am

bugman wrote in Mon May 27, 2019 9:17 am:But as a copyrighted sound was used as a starting point

I can't find a statement in this thread as to whether or not "a copyrighted sound was used as a starting point".
OSM buildings for LOWI, EDDC
Custom scenery for VHXX YMML
Edit .stg via the FG Object Placement Tool
radi
 
Posts: 659
Joined: Mon Aug 25, 2008 5:24 pm
Location: YMML, EDDC

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby bugman » Mon May 27, 2019 10:22 am

See:

ghedger42 wrote in Sat May 18, 2019 7:28 am:...for all you know, it may be something I or a friend recorded)...


Again such unclear language and obfuscation of the sound source does not instil confidence! If there really is no copyright issues, why is the source not clearly stated? Anyway, analytic harmonic analysis relies on an input data source - in this case a sound recording - as far as I am aware.

Regards,
Edward
bugman
Moderator
 
Posts: 1808
Joined: Thu Mar 19, 2015 10:01 am
Version: next

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby radi » Mon May 27, 2019 10:35 am

I've read that. Sure, its unclear, vague, sounds a bit fishy and doesn't build confidence. But that doesn't make the source copyrighted. Would make it "unknown origin" in my book.
OSM buildings for LOWI, EDDC
Custom scenery for VHXX YMML
Edit .stg via the FG Object Placement Tool
radi
 
Posts: 659
Joined: Mon Aug 25, 2008 5:24 pm
Location: YMML, EDDC

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby AndersG » Mon May 27, 2019 10:45 am

If we knew the initial sound sample (if there is one) was ok to redistribute under GPLv2+ then there would be no particular need to decide if the generated sounds are derived works or not to distribute them under GPLv2+.
Callsign: SE-AG
Aircraft (uhm...): Submarine Scout, Zeppelin NT, ZF Navy free balloon, Nordstern, Hindenburg, Short Empire flying-boat, ZNP-K, North Sea class, MTB T21 class, U.S.S. Monitor, MFI-9B, Type UB I submarine, Gokstad ship, Renault FT.
AndersG
 
Posts: 2524
Joined: Wed Nov 29, 2006 10:20 am
Location: Göteborg, Sweden
Callsign: SE-AG
OS: Debian GNU Linux

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby Alant » Mon May 27, 2019 11:33 am

From the OP´s posts I think that we can assume that the source is not GPL.
My view is that if the simulated sound is made by tuning waveforms so that they sound similar to the original, then this is no worse than making a texture map that that looks similar to a picture seen in a book or on the internet.
If it s done by measuring the amplitude and frequency of the sound spectrum, then this is comparable to producing a texture map by image processing, tracing or editing the original picture.
The first method is widely done in FG and is probably safe to do . The second is probably not.
Alant
 
Posts: 1219
Joined: Wed Jun 23, 2010 6:58 am
Location: Portugal
Callsign: Tarnish99
Version: latest Git
OS: Windows 10/11

Re: Improved Engine Sound work done: P&W JT3D on Boeing 707

Postby bugman » Mon May 27, 2019 1:10 pm

radi wrote in Mon May 27, 2019 10:35 am:...But that doesn't make the source copyrighted. Would make it "unknown origin" in my book.


Everything should be assumed to be copyrighted. By law in almost all countries in the world, you automatically have copyright on things you create. Some countries give you greater protection if you register your copyright (the US for example), but you nevertheless own the copyright by default. You have to publicly disclaim the copyright to make it public domain, and hence not be copyrighted.

Regards,
Edward
bugman
Moderator
 
Posts: 1808
Joined: Thu Mar 19, 2015 10:01 am
Version: next

PreviousNext

Return to Aircraft

Who is online

Users browsing this forum: No registered users and 11 guests