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

Children's speech recognition is a vital, yet largely overlooked domain whenbuilding inclusive speech technologies. The major challenge impeding progressin this domain is the lack of adequate child speech corpora; however, recentadvances in self-supervised learning have created a new opportunity forovercoming this problem of data scarcity. In this paper, we leverageself-supervised adult speech representations and use three well-known childspeech corpora to build models for children's speech recognition. We assess theperformance of fine-tuning on both native and non-native children's speech,examine the effect of cross-domain child corpora, and investigate the minimumamount of child speech required to fine-tune a model which outperforms astate-of-the-art adult model. We also analyze speech recognition performanceacross children's ages. Our results demonstrate that fine-tuning withcross-domain child corpora leads to relative improvements of up to 46.08% and45.53% for native and non-native child speech respectively, and absoluteimprovements of 14.70% and 31.10%. We also show that with as little as 5 hoursof transcribed children's speech, it is possible to fine-tune a children'sspeech recognition system that outperforms a state-of-the-art adult modelfine-tuned on 960 hours of adult speech.

Improving Children's Speech Recognition by Fine-tuning Self-supervised Adult Speech Representations