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

In low-bitrate speech coding, end-to-end speech coding networks aim to learncompact yet expressive features and a powerful decoder in a single network. Achallenging problem as such results in unwelcome complexity increase andinferior speech quality. In this paper, we propose to separate therepresentation learning and information reconstruction tasks. We leverage anend-to-end codec for learning low-dimensional discrete tokens and employ alatent diffusion model to de-quantize coded features into a high-dimensionalcontinuous space, relieving the decoder's burden of de-quantizing andupsampling. To mitigate the issue of over-smooth generation, we introducemidway-infilling with less noise reduction and stronger conditioning. Inablation studies, we investigate the hyperparameters for midway-infilling andlatent diffusion space with different dimensions. Subjective listening testsshow that our model outperforms the state-of-the-art at two low bitrates, 1.5and 3 kbps. Codes and samples of this work are available on our webpage.

Generative De-Quantization for Neural Speech Codec via Latent Diffusion