WaveGrad: Estimating Gradients for Waveform Generation
This paper introduces WaveGrad, a conditional model for waveform generation which estimates gradients of the data density. The model is built on prior work on score matching and diffusion probabilistic models. It starts from a Gaussian white noise signal and iteratively refines the signal via a gradient-based sampler conditioned on the mel-spectrogram. WaveGrad offers a natural way to trade inference speed for sample quality by adjusting the number of refinement steps, and bridges the gap between non-autoregressive and autoregressive models in terms of audio quality. We find that it can generate high fidelity audio samples using as few as six iterations. Experiments reveal WaveGrad to generate high fidelity audio, outperforming adversarial non-autoregressive baselines and matching a strong likelihood-based autoregressive baseline using fewer sequential operations. Audio samples are available at this https URL.
WAVE G RAD : E STIMATING G RADIENTS FOR WAVEFORM G ENERATION Nanxin Chen∗ Johns Hopkins University, Center for Language and Speech Processing bobchennan@jhu.edu Yu Zhang†, Heiga Zen, Ron J. Weiss, Mohammad Norouzi, William Chan† Google Research, Brain Team…
related reading
- Generating music in the waveform domain – Sander Dielemansander.ai
- What are Diffusion Models? | Lil'Loglilianweng.github.io
- Yang Songyang-song.net
- Diffusion is spectral autoregression – Sander Dielemansander.ai
- What are Diffusion Models?lilianweng.github.io
- Diffusion models from scratchchenyang.co
- GANSynth: Making music with GANsmagenta.tensorflow.org
- WaveNetdeepmind.com
- Woosh: A Sound Effects Foundation Modelarxiv.org
- Generative Modeling by Estimating Gradients of the Data Distribution | Yang Songyang-song.github.io
- Diffusion model - Wikipediaen.wikipedia.org
- https://arxiv.org/pdf/2006.11239arxiv.org