Audio Super Resolution using Neural Networks

Abstract

We introduce a new audio processing technique that increases the sampling rate of signals such as speech or music using deep convolutional neural networks. Our model is trained on pairs of low and high-quality audio examples; at test-time, it predicts missing samples within a low-resolution signal in an interpolation process similar to image super-resolution. Our method is simple and does not involve specialized audio processing techniques; in our experiments, it outperforms baselines on standard speech and music benchmarks at upscaling ratios of 2x, 4x, and 6x. The method has practical applications in telephony, compression, and text-to-speech generation; it demonstrates the effectiveness of feed-forward convolutional architectures on an audio generation task.

Presented at the 5th International Conference on Learning Representations (ICLR) 2017, Workshop Track, Toulon, France

References in corpus (1)

• SoundNet: Learning Sound Representations from Unlabeled Video

Cited by in corpus (8)

• Stochastic Restoration of Heavily Compressed Musical Audio using Generative Adversarial Networks
• VoiceFixer: Toward General Speech Restoration with Neural Vocoder
• NU-GAN: High resolution neural upsampling with GAN
• Super-Resolution Perception for Industrial Sensor Data
• SplitSR: An End-to-End Approach to Super-Resolution on Mobile Devices
• Using Cyclic Noise as the Source Signal for Neural Source-Filter-based Speech Waveform Model
• Phase-aware music super-resolution using generative adversarial networks
• An investigation of pre-upsampling generative modelling and Generative Adversarial Networks in audio super resolution