Second Mesh Secures Ufi VocTech Trust Funding

ESOL (English for Speakers of Other Languages) speaking and listening assessments are currently recorded and reviewed manually by teachers. The process is time-consuming and labour-intensive, and marking at the scale of a whole cohort can lead to inconsistent feedback. Because these assessments place a high demand on teacher time, they happen infrequently, and learners often wait too long for feedback. The delay makes it harder for students to improve their language skills, affecting confidence and progression. With over 160,000 ESOL learners in England, there is clear demand for more timely and effective feedback to support progression into further education, vocational training or employment.

Through funding from Ufi VocTech Trust, Second Mesh will develop and pilot an AI video assessment and feedback tool for ESOL speaking and listening tests. The tool will:

• Analyse recorded assessment videos using teacher-defined mark schemes.
• Generate structured feedback that follows the teacher's mark scheme.
• Keep teachers in control: they review and refine the AI's output before it is shared with learners.
• Give learners on-demand feedback through a chatbot interface.
• Produce insights across groups of learners to support moderation and identify common skills gaps.

Our aim is to help ESOL teachers deliver effective assessment feedback to their students, while reducing the time it takes to review a cohort's speaking and listening assessments.

Beyond the project, the elements which make up the AI video assessment and feedback tool will be repurposed to support a broader set of challenges that centre on the use of audio and video in teaching, learning and assessment.

The project is funded by Ufi, an independent charity focused on unlocking the full potential of technology to get adults learning.