Deep Audio Priors Emerge From Harmonic Convolutional Networks
Convolutional neural networks (CNNs) excel in image recognition and generation. Among many efforts to explain their effectiveness, experiments show that CNNs carry strong inductive biases that capture natural image priors. Do deep networks also have inductive biases for audio signals? In this paper, we empirically show that current network architectures for audio processing do not show strong evidence in capturing such priors. We propose Harmonic Convolution, an operation that helps deep networks distill priors in audio signals by explicitly utilizing the harmonic structure within. This is done by engineering the kernel to be supported by sets of harmonic series, instead of local neighborhoods for convolutional kernels. We show that networks using Harmonic Convolution can reliably model audio priors and achieve high performance in unsupervised audio restoration tasks. With Harmonic Convolution, they also achieve better generalization performance for sound source separation.
Code (0)
등록된 구현이 없습니다.
Methods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Deep Audio Waveform Prior
Convolutional neural networks contain strong priors for generating natural looking images [1]. These priors enable image denoising, super resolution, and inpainting in an unsupervised manner. Previous attempts to demonst…
Audio inpaintingAudio Source SeparationDenoisingImage Denoising+1Efficient Learning of Harmonic Priors for Pitch Detection in Polyphonic Music
Automatic music transcription (AMT) aims to infer a latent symbolic representation of a piece of music (piano-roll), given a corresponding observed audio recording. Transcribing polyphonic music (when multiple notes are …
Music TranscriptionOn the Design of Deep Priors for Unsupervised Audio Restoration
Unsupervised deep learning methods for solving audio restoration problems extensively rely on carefully tailored neural architectures that carry strong inductive biases for defining priors in the time or spectral domain.…
Audio DenoisingDenoisingAcoustic Scene Classification Using Bilinear Pooling on Time-liked and Frequency-liked Convolution Neural Network
The current methodology in tackling Acoustic Scene Classification (ASC) task can be described in two steps, preprocessing of the audio waveform into log-mel spectrogram and then using it as the input representation for C…
Acoustic Scene ClassificationGeneral ClassificationInformation RetrievalMusic Information Retrieval+2Synthnet: Learning synthesizers end-to-end
Learning synthesizers and generating music in the raw audio domain is a challenging task. We investigate the learned representations of convolutional autoregressive generative models. Consequently, we show that mappings …