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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Neural models are known to be over-parameterized, and recent work has shownthat sparse text-to-speech (TTS) models can outperform dense models. Although aplethora of sparse methods has been proposed for other domains, such methodshave rarely been applied in TTS. In this work, we seek to answer the question:what are the characteristics of selected sparse techniques on the performanceand model complexity? We compare a Tacotron2 baseline and the results ofapplying five techniques. We then evaluate the performance via the factors ofnaturalness, intelligibility and prosody, while reporting model size andtraining time. Complementary to prior research, we find that pruning before orduring training can achieve similar performance to pruning after training andcan be trained much faster, while removing entire neurons degrades performancemuch more than removing parameters. To our best knowledge, this is the firstwork that compares sparsity paradigms in text-to-speech synthesis.

EPIC TTS Models: Empirical Pruning Investigations Characterizing Text-To-Speech Models