Who we are
English Language iTutoring (ELiT) is a tech-transfer company founded in 2014 as a joint venture with Cambridge University Press and Cambridge Assessment. In 2019, it was acquired by Cambridge University Press and Assessment (CUP&A).
As part of the CUP&A family, ELiT operationalises Machine Learning (AI) research from the Cambridge University Institute for Automated Language Teaching and Assessment (ALTA) to provide automated assessment capabilities (automarking and automated feedback) for extended English learner text and speech. ALTA is a virtual institute which investigates new ways of using technology to enhance language learning and to develop cutting-edge approaches to assessment and feedback which will benefit English learners and teachers worldwide. These capabilities can be accessed through ELiT's APIs by CUP&A’s learning and assessment products.
ELiT is a remote, agile team consisting of 11 software engineers and an engineering manager, a senior data scientist, a product owner, a technical architect, and a data annotation and EFL product specialist. We build and maintain ground-breaking English Language Learning applications that are used by hundreds of thousands of learners and teachers worldwide, and our automated language assessment technology delivers exam results to candidates without delay. Our innovative annotation tool is used to enrich learner data and improve the performance of our machine learning pipeline and our products.
Our mission
ELiT's mission is:
- to operate secure, efficient and highly scalable services enabling automated feedback and assessment for Cambridge's English language learning and assessment products
- to collect and enrich, through annotation, large amounts of diverse data from English language learners and teachers to support cutting-edge research at Cambridge University into the use of ethical AI to support English language learning and teaching, fuelling a continuous cycle of improvement in existing and future products and APIs
- to continue and build on our early ground-breaking AI productisation to support CUP&A as a leader in the field.
Our Products and Services
Text and Speech auto-marking APIs
Our Text and Speech APIs are core to what we do. They provide automatic assessment capabilities to the products we build and to those within the wider CUP&A organisation. They enable API clients to submit extended text and speech for automatic assessment by a range of machine learning models and to receive the results. They are designed to be resilient and to scale automatically in response to demand, so that assessment results are returned in a timely manner.
The &Improves
We believe that the best way to improve at any skill is through practice and timely, actionable feedback. We develop and operate three Integrated Learning and Assessment products for English language learners – Write & Improve, Speak & Improve and Read & Improve, all at different stages of maturity and under constant development. These offer learners intensive, single-skill practice with automatic assessment and feedback designed to encourage them to learn from their mistakes and try again.
Write & Improve also offers two premium products:
- +Test Zone for learner exam practice (B2 First and IELTS)
- +Class View for teachers to set, mark and manage students’ writing practice.
Through these products, ELiT gathers diverse, up-to-the-minute data and insights, which help to support further learning and assessment research by ALTA and other research partners. This creates a continuous cycle of learning and improvement which is passed back into these and other CUP&A learning tools.
Data annotation
Continuing CUP&A's long-standing commitment to data-led learning and assessment, ELiT provides CUP&A and ALTA with innovative, bespoke data-annotation services through a team of specially trained EFL teachers and examiners, teacher trainers, linguists and lexicographers. Using our unique spoken- and written-language annotation tool, Manatee, we ensure that the human is always in the loop when operationalising AI solutions for automated assessment and feedback for English learners.
Manatee is used for written error annotation, automarker assessment and validation, analytic scoring of speech, Automatic Speech Recognition training on errorful learner speech, and for multiple projects to advance annotation capabilities aimed at developing pedagogically sound, actionable feedback for learners on their extended speech and writing. The tool is designed to be flexible and scalable, growing with the needs of the organisation.
CUP&A non-commercial research datasets
Public Research Dataset Policy
ELiT, as part of Cambridge University Press & Assessment (CUP&A), is committed to contributing to society through the pursuit of education, learning and research at the highest international levels of excellence. CUP&A funds and supports research undertaken by the Cambridge University Institute for Automated Language Teaching & Assessment (ALTA), utilising the expertise of both the Computer Science & Technology Department and the Engineering Department to investigate innovative ways of using technology to enhance English language education. We believe that by making data publicly available for academic research, we will encourage further research based on these datasets.
