A Unified Neural Architecture For Instrumental Audio Tasks

Abstract

Within Music Information Retrieval (MIR), prominent tasks -- including pitch-tracking, source-separation, super-resolution, and synthesis -- typically call for specialised methods, despite their similarities. Conditional Generative Adversarial Networks (cGANs) have been shown to be highly versatile in learning general image-to-image translations, but have not yet been adapted across MIR. In this work, we present an end-to-end supervisable architecture to perform all aforementioned audio tasks, consisting of a WaveNet synthesiser conditioned on the output of a jointly-trained cGAN spectrogram translator. In doing so, we demonstrate the potential of such flexible techniques to unify MIR tasks, promote efficient transfer learning, and converge research to the improvement of powerful, general methods. Finally, to the best of our knowledge, we present the first application of GANs to guided instrument synthesis.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyspratley2019a

Related papers

• Gansynth: Adversarial Neural Audio Synthesis (2019)0.00
• Melgan: Generative Adversarial Networks For Conditional Waveform Synthesis (2019)0.00
• Adversarial Audio Synthesis (2018)0.00
• Bandwidth Extension On Raw Audio Via Generative Adversarial Networks (2019)0.00
• Bigwavgan: A Wave-to-wave Generative Adversarial Network For Music Super-resolution (2023)0.00
• Midi-sandwich: Multi-model Multi-task Hierarchical Conditional VAE-GAN Networks For Symbolic Single-track Music Generation (2019)0.00
• Waveform Generation For Text-to-speech Synthesis Using Pitch-synchronous Multi-scale Generative Adversarial Networks (2018)8.35
• Neural Music Synthesis For Flexible Timbre Control (2018)9.92