Wow, this is seriously impressive. The author makes a very bold claim: "Get ready for a revolutionary leap in sound synthesis technology that will render all e

Wow, this is seriously impressive. The author makes a very bold claim: "Get ready for a revolutionary leap in sound synthesis technology that will render all existing synthesizers obsolete." I have to admit, when I heard that I was really skeptical. "With the power of SynthGPT, you can now create any sound imaginable, simply by typing a description." But, how do you even describe a sound? (Well, look at the descriptions in the video...) I've done a bit of music production myself, and I always wanted to create the sound of wind chimes on a synthesizer. The really "delicate", airy wind chimes. But no synth could produce something light enough, even physical modeling synths like https://www.applied-acoustics.com/chromaphone-3/ https://www.applied-acoustics.com/chromaphone-3/ which are supposedly designed for this. And these synths are very expensive. I typed in the prompt in SynthGPT "Emulate a delicate, airy wind-chime on a quiet summer day." and I have to say the output was perfect. I have no idea how it got my computer to make that sound when every other "Wind Chimes Demo" couldn't. I also tried getting SynthGPT to emulate various real synthesizers, and compared to live recordings on YouTube: the comparison is strikingly similar. To the point where I don't know why someone would spend money on a real synth anymore when they have SynthGPT. Most GPT products, even when impressive, are simply not ready for real-world or commercial use. But I seriously believe SynthGPT could be used for a real song and I wouldn't even notice.