English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Denoising diffusion probabilistic models (diffusion models for short) requirea large number of iterations in inference to achieve the generation qualitythat matches or surpasses the state-of-the-art generative models, whichinvariably results in slow inference speed. Previous approaches aim to optimizethe choice of inference schedule over a few iterations to speed up inference.However, this results in reduced generation quality, mainly because theinference process is optimized separately, without jointly optimizing with thetraining process. In this paper, we propose InferGrad, a diffusion model forvocoder that incorporates inference process into training, to reduce theinference iterations while maintaining high generation quality. Morespecifically, during training, we generate data from random noise through areverse process under inference schedules with a few iterations, and impose aloss to minimize the gap between the generated and ground-truth data samples.Then, unlike existing approaches, the training of InferGrad considers theinference process. The advantages of InferGrad are demonstrated throughexperiments on the LJSpeech dataset showing that InferGrad achieves bettervoice quality than the baseline WaveGrad under same conditions whilemaintaining the same voice quality as the baseline but with $3$x speedup ($2$iterations for InferGrad vs $6$ iterations for WaveGrad).

InferGrad: Improving Diffusion Models for Vocoder by Considering Inference in Training