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 (DDPMs) have recently achievedleading performances in many generative tasks. However, the inherited iterativesampling process costs hindered their applications to speech synthesis. Thispaper proposes FastDiff, a fast conditional diffusion model for high-qualityspeech synthesis. FastDiff employs a stack of time-aware location-variableconvolutions of diverse receptive field patterns to efficiently model long-termtime dependencies with adaptive conditions. A noise schedule predictor is alsoadopted to reduce the sampling steps without sacrificing the generationquality. Based on FastDiff, we design an end-to-end text-to-speech synthesizer,FastDiff-TTS, which generates high-fidelity speech waveforms without anyintermediate feature (e.g., Mel-spectrogram). Our evaluation of FastDiffdemonstrates the state-of-the-art results with higher-quality (MOS 4.28) speechsamples. Also, FastDiff enables a sampling speed of 58x faster than real-timeon a V100 GPU, making diffusion models practically applicable to speechsynthesis deployment for the first time. We further show that FastDiffgeneralized well to the mel-spectrogram inversion of unseen speakers, andFastDiff-TTS outperformed other competing methods in end-to-end text-to-speechsynthesis. Audio samples are available at \url{https://FastDiff.github.io/}.

FastDiff: A Fast Conditional Diffusion Model for High-Quality Speech Synthesis