Forum Replies Created
-
Posts
-
Contrary to popular belief, China offers fairly robust protection of intellectual property, including music.
The 2021 report from IFPI showed that 97% of Chinese consumers used licensed music content. The global average is below 70%. Also, 61% of Chinese consumers were subscribed to at least one paid service.
The 3rd Amendment to the Chinese Copyright Act, enacted in 2021, increased copyright protections on audio-visual works.
China had no tradition of intellectual property law under the Dynastic System before the 1920s, but IP law in China has developed rapidly since the 90s. They now file twice as many international patents as the US, and created a dedicated court system to deal with IP infringement.
I’m not sure how they will deal with AI, but I wouldn’t assume it will be worse than any other country.
For acoustic dampening, I use batts of Rockwool insulation (generically called mineral wool). Same as in the videos above, except I used batts instead of rigid sheets, because that’s what they carry at Home Depot. Rockwool works much better than foam, and costs less than commercial kits. Otherwise, my panels are built in a similar way to the video.
I also have a monstrous freestanding Rockwool gobo. It’s a similar idea to the panels in the video, but enormous and built with 2x4s. If you want something that will destroy all room sound near a mic, this would do it. It’s about 4’x6′ and framed similar to an interior wall of a house. It’s crammed 2 layers deep with mineral wool. Ordinary fabric wouldn’t contain the insulation, so I used a canvas painter’s tarp with some wood braces on the outside of the tarp to hold it all in. It’s… slightly overkill. LOL
Even in my treated studio, you can tell the difference if you point one ear at it. It mostly stays tucked out of the way as a bass trap. Setting up 2 of these at a right angle wouldn’t be very far removed from an isolation booth. Since it is moveable, it offers a lot of flexibility.
You could downscale that idea and use smaller panels set up as baffles behind a mic.
But I will second the recommendation to use a dynamic mic placed close, unless it’s something really delicate and you absolutely must have a condenser.
Also, don’t rule out the possibility of other rooms in the house. I’ve gotten pretty good results occasionally by using the living room as a live room. A couple of extra mic cables and a headphone extension cable don’t cost much.
If I have an existing relationship with a publisher, I will give them the courtesy of waiting for a yes or no answer (within reason). I understand they are busy and they have backlogged tracks. That’s a normal part of the business. I try to keep track of their normal turnaround times so I know roughly how it will take. If I don’t hear from them within a couple of weeks after I expect to, I would follow up and make sure they received my submission. I extend them the professional courtesy of setting aside the cues I submit until they are able to review them so I do not waste their time. In return, I expect the professional courtesy of a yes or no response when they review them. If I had a relationship with a library and they started ignoring my submissions altogether, I wouldn’t want to continue that relationship. Fortunately, this has never happened to me.
For new libraries with whom I have no relationship, I now give 3 or 4 business days. It’s a compromise approach. I also used to wait 2 weeks, but I don’t anymore. A library can’t reasonably expect a composer with whom they have no relationship to withhold inventory for them. But I also do not want to waste people’s time, and I want to avoid the embarrassment of telling someone the album I sent yesterday is already unavailable. So I usually give them 3 or 4 business days before I move on to the next library. First come, first served.
I would love to be able to give more time than that, but my ReelCrafter stats tell me that somewhere around 95% of the cold pitches I send never get opened at all. If I send to 13 libraries and wait 2 weeks for each, that‘s 6 months of sitting on tracks that could have already been published with someone else, and it’s possible that nobody will have even clicked on the link to my playlist. Most people who run libraries are kind and reasonable people, and I‘m sure they understand that. And, of course, I’m always happy to make another album for them.
Well, I hope I’m right, but I have been wrong from time to time in the past. 😉 Maybe I’m whistling past the graveyard.
I personally just don’t see a market for AI music any time soon. Even if it can create music of the same quality as stock libraries, I don’t see why someone wouldn’t just continue to use stock libraries. Stock music is very saturated. Why mess with a computer program (or pay someone to mess with a computer program) to create a track that can already be licensed for $3?
The same thing is already happening with AI images. It’s certainly fun to mess around with AI art. But if I need an image of a rolling landscape, there are thousands of existing stock photos available for free or for pennies. I don’t see people using AI images commercially, as far as I can tell anyway.
Production music isn’t very much different. A lot of production music is already being placed with no sync fees. If I’m a music editor, where’s my incentive to create AI music when there are millions of tracks already available that I’m placing at No up-front cost? The networks are paying for PRO licenses anyway. Unless they ditch all their existing programming that has any music written by humans, they won’t save any money by using AI music, because they still have to maintain their PRO licenses. They would have to ditch the PRO model altogether, which would mean ditching every note of music written by humans. I just don’t see that happening en masse any time soon.
Sure, programmers may develop the technical ability to create something resembling stock music, but it doesn’t have a market that isn’t already saturated.
AI music is also far more complex than AI images. An image is a 2D array of pixels, but music is like thousands of 2D images sequenced in a coherent way across a time domain. You can certainly use AI to create some MIDI notes, but that’s very different from a finished composition mixed and mastered by a skilled composer. You could hand over your MIDI data to a non-composer, and tell them exactly which instruments to use, and they wouldn’t be able to replicate the quality of your work.
In short, I think market saturation from more and more composers writing music is a much bigger threat to the income of composers, but I don’t have the right to decide who gets to create music, so all I can do is keep trying to get better.