Hello, the faculty of music at my university in Norway has been granted a lot of money to focus on the use of technology in music and music education. I have been accepted into a position focusing on AI in music. I am first and foremost a musician, so i am working on getting a fundamental grasp on machine learning.

I was wondering if you know of any website or software that can be used to train a signal processing model. The idea being that i can give an unprocessed signal and a processed signal and have mode learn the differences. Like a guitar playing without a distortion pedal and with one. Thank you very much. If there are any misconceptions in my question please help educate me.

  • qalis@alien.topBEnglish
    1·
    1 year ago

    You will probably get better answers at r/LearnMachineLearning. You need some basic neural networks courses for beginning. Then e.g. HuggingFace’s audio processing course, it’s short and high-level, but will be a nice intro. In general you will focus on convolutional networks (CNNs) and processing audio either as 1D signals or 2D images (spectrograms).