引言

在这里放一些自己感兴趣的方向的论文以及AI音频音乐领域的相关产出。

应用

  1. Vocal Remover audio source separation
  2. UVR5 audio source separation
  3. DrumBot: your real-time ML drummer
  4. Muzic music understanding and generation
  5. Basic Pitch transcription
  6. Audiocraft provides the code and models for MusicGen

视频

  1. Music + AI Reading Group

API/数据集/工具

  1. Magenta
  2. EGFxSet (Electric Guitar Effects Dataset)
  3. Web Audio API 基于Javascript的Web API
  4. MidiTok convert MIDI files into tokens

论文/研究

音频分离 Audio Separation

  1. Distortion Audio Effects: Learning How to Recover the Clean Signal
  2. Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation

自动混音 Automatic Mixing

  1. Ten Years of Automatic Mixing(2017)
  2. A Deep Learning Approach to Intelligent Drum Mixing with the Wave-U-Net
  3. Style Transfer of Audio Effects with Differentiable Signal Processing
  4. Deep Learning for Black-Box Modeling of Audio Effects
  5. Modeling Plate and Spring Reverberation Using A Dsp-Informed Deep Neural Network

音乐分类 Music Classification

  1. Genre Classification of Electronic Dance Music Using Spotify’s Audio Analysis 这个是用简单的机器学习分类方法进行实践，有源码和讲解，特别适合入门。

识别

  1. Computational Analysis of Sound Scenes and Events
  2. Accompanying Website for Synthesizer Sound Matching with Differentiable DSP

转录Transcription

  1. Automatic Music Transcription: An Overview
  2. MT3: Multi-Task Multitrack Music Transcription
  3. DrummerNet – Deep Unsupervised Drum Transcription

音乐生成 Music Generation

  1. Music Modeling and Music Generation with Deep Learning
  2. ChordAL - A chord-based approach for AI music generation
  3. MODELING THE RHYTHM FROM LYRICS FOR MELODY GENERATION OF POP SONG