Kate Knill

Take part in the Speak & Improve Challenge 2025!

In connection with the ISCA SLaTE 2025 Workshop and Cambridge University Press and Assessment, we are happy to introduce the Speak & Improve Challenge 2025 to the speech and language learning community. Our goal is to advance the technology in the field of spoken language assessment and feedback by making available a new rich dataset and proposing a variety of associated tasks.

The challenge offers a unique opportunity with the pre-release of the Speak & Improve dataset from Cambridge University Press & Assessment. This dataset is derived from the Cambridge English Speak & Improve L2 (second language) English speaking practice tool and contains annotated recordings of a wide variety of L2 English learner speech on open (spontaneous) speaking tasks.

Key dates:

• Registration for Challenge: open now
• Training data and development data available: now
• Evaluation data and opening of Challenge submission site: 26th March 2025
• Closing of submission site: 2nd April 2025
• Announcement of results: 9th April 2025
• SLaTE paper submision deadline: 22nd May 2025

Kate Knill is a Principal Research Associate in the Speech Research Group of the Machine Intelligence Laboratory working primarily on the ALTA Institute Spoken Language Technology Project, of which she is the Principal Investigator. She was the Information Engineering (Div F) Athena SWAN champion. Learn more about CUED's Engineering Diversity here. Prior to joining the lab, she was an Assistant Managing Director of Toshiba Research Europe Ltd, Cambridge Research Lab, leading the Speech Technology Group she established in 2002.

A brief biography is available online.

UK and Ireland Speech (UKIS) 2024
UK Speech 2017

Research interests

• Large vocabulary continuous speech recognition
• Statistical parametric speech synthesis
• Multi-language speech recognition and speech synthesis
• Assistive speech technology

Research Projects

• ALTA - Spoken Language Research @ CUED funded by Cambridge University Press & Assessment - local web page
• 4th year (IIB) projects - local web page

• Detection of speech disorders in children's speech - project led by Avin Mirawdeli, Faculty of Education, - local web page
• MATERIAL (IARPA funded) - local web page
• BABEL (IARPA funded project) - local web page
• RATS (DARPA funded project) - local web page

Presentations

• Foundation Models in Spoken Language Processing: Time to go home or make hay?
  Observations from automated language learning and assessment
  Keynote at UK Speech 2023 (Jun 2023)
• Automated Assessment and Feedback: the Role of Spoken Grammatical Error Correction
  Keynote at ISCSLP 2022 (Dec 2022)
• Applying Deep Learning in Non-native Spoken English Assessment
  Keynote at APSIPA (Nov 2019)
• Automatic Grammatical Error Detection of Non-native Spoken Learner English
  ICASSP 2019 (May 2019)
• Towards Learning Orientated Assessment for Non-native Spoken English
  Seminar at Sheffield University (March 2019)
• Challenges for AI in Spoken Communication
  British Embassy UK-Japan AI Workshop (March 2017)
• Machine Learning of Level and Progression in Second/Additional Language Spoken English
  Seminar at University of Birmingham (May 2016)
• Machine Learning of Level and Progression in Spoken EAL
  Cambridge English Profile (Feb 2016)
• (Deep) Neural Networks for Speech Processing: Part 1, Part 2
  Tutorial at DREAMS Summer School (Sep 2015)
• Language Independent and Unsupervised Acoustic Models for Speech Recognition and Keyword Spotting
  Interspeech 2014 (Sep 2014)
• Investigation of multilingual speech-to-text systems for use in spoken term detection
  Seminar at CUED (Apr 2014)
• Investigation of multilingual speech-to-text systems for use in spoken term detection
  Seminar at University of Edingburgh (Feb 2014)

Contact Information