Thyroid ultrasound diagnosis improvement via multi-view self-supervised learning and two-stage pre-training

no code implementations • 18 Feb 2024 • Jian Wang, Xin Yang, Xiaohong Jia, Wufeng Xue, Rusi Chen, Yanlin Chen, Xiliang Zhu, Lian Liu, Yan Cao, Jianqiao Zhou, Dong Ni, Ning Gu

In this study, we proposed a multi-view contrastive self-supervised method to improve thyroid nodule classification and segmentation performance with limited manual labels.

Classification Segmentation +1

Resolving Transcription Ambiguity in Spanish: A Hybrid Acoustic-Lexical System for Punctuation Restoration

no code implementations • 5 Feb 2024 • Xiliang Zhu, Chia-Tien Chang, Shayna Gardiner, David Rossouw, Jonas Robertson

Punctuation restoration is a crucial step after Automatic Speech Recognition (ASR) systems to enhance transcript readability and facilitate subsequent NLP tasks.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +4

Extracting Similar Questions From Naturally-occurring Business Conversations

no code implementations • 3 Jun 2022 • Xiliang Zhu, David Rossouw, Shayna Gardiner, Simon Corston-Oliver

Pre-trained contextualized embedding models such as BERT are a standard building block in many natural language processing systems.

Sentence

Punctuation Restoration in Spanish Customer Support Transcripts using Transfer Learning

no code implementations • DeepLo 2022 • Xiliang Zhu, Shayna Gardiner, David Rossouw, Tere Roldán, Simon Corston-Oliver

Automatic Speech Recognition (ASR) systems typically produce unpunctuated transcripts that have poor readability.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +5