This project uses tensorflow/keras to emulate the sound of real guitar amplifiers and pedals on wav files. The model is trained on about three minutes of input/output guitar audio, and takes a few minutes to train. For example, you can record a $1000 amp and use this code to create a model to apply that sound to other wav files.
The purpose of this project is to improve on the previous WaveNet model built for the same task. Using LSTM is orders of magnitude faster and more accurate for emulating guitar signals than WaveNet. A real time guitar plugin exists for the WaveNet implementation, and a plugin for this the LSTM model is currently in work. Compare to Neural DSP’s Quad Cortex neural capture feature, which uses a similar technique on their $1600 floor modeler.
Hey, I just wanted to say this is awesome and if you need somebody to help test it I'd love to help.
I'm a software engineer, but also a big guitarist. The WaveNet stuff is really interesting, but I'm not a DSP/ML expert. I could help you:
* Model a Dr. Z Maz 38 MkII, 5w Fender Tweed Champ, Ampeg SCR-DI preamp/DI
* Provide raw samples of: Gibson Les Paul 1958 Reissue (R8), Fender Telecaster (1955 type partscaster), Fender Stratocaster (1962 "thin skin" reissue), Gibson 1963 SG Junior Reissue, Fender 1958 Precision Bass Reissue, Taylor 814ce acoustic guitar
* Get the code installed on my local, test whatever you need help testing, etc.
* Testing and feedback on the real-time plugin i.e. using a VST/AU in Logic, testing latency, testing sound against in-room amp sound, etc.
I'd also be interested in figuring out how to get the real-time plugin you're developing into a pedal enclosure. I have a friend who is a top notch silicon valley hardware/power engineer, and we've been toying with the Strymon Iridium, Impulse Responses, analog pedal building, etc. It'd be interesting to use some guerrilla marketing, sell a few pedals on The Gear Page, and see what people think.
benstandefer (at) <google's email service> (dot) com
This is super cool! Congratulations! I am doing sth in the same space but related to recording - check out Band Studio app. My Show HN post is couple of items below yours. I am thinking how those two things could play together...
That is a really cool idea, I’m in a local band and we record from our phone if we want a quick way to remember how we played something. Your Band Studio app would add a whole new level to that, especially considering how hard it is to get everyone together in one place. If you wanted to apply a certain sound to the tracks you could use something like what I made on guitar/bass, or possibly a microphone model, in the same way you’d use in on a traditional DAW.
The purpose of this project is to improve on the previous WaveNet model built for the same task. Using LSTM is orders of magnitude faster and more accurate for emulating guitar signals than WaveNet. A real time guitar plugin exists for the WaveNet implementation, and a plugin for this the LSTM model is currently in work. Compare to Neural DSP’s Quad Cortex neural capture feature, which uses a similar technique on their $1600 floor modeler.