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

As a key component of automated speech recognition (ASR) and the front-end intext-to-speech (TTS), grapheme-to-phoneme (G2P) plays the role of convertingletters to their corresponding pronunciations. Existing methods are either slowor poor in performance, and are limited in application scenarios, particularlyin the process of on-device inference. In this paper, we integrate theadvantages of both expert knowledge and connectionist temporal classification(CTC) based neural network and propose a novel method named LiteG2P which isfast, light and theoretically parallel. With the carefully leading design,LiteG2P can be applied both on cloud and on device. Experimental results on theCMU dataset show that the performance of the proposed method is superior to thestate-of-the-art CTC based method with 10 times fewer parameters, and evencomparable to the state-of-the-art Transformer-based sequence-to-sequence modelwith less parameters and 33 times less computation.

LiteG2P: A fast, light and high accuracy model for grapheme-to-phoneme conversion