Speak & Improve Corpus 2025: an L2 English Speech Corpus for Language Assessment and Feedback

TL;DR

The paper presents Speak & Improve Corpus 2025, a large public dataset of L2 English speech with holistic proficiency scores and granular grammatical-error annotations across CEFR levels from $A2$ to $C1$, addressing the scarcity of high-quality labeled data for spoken language processing. It describes data collection from the Speak & Improve platform and a three-phase annotation pipeline delivering holistic scores, ASR-derived transcripts with error tagging, and reference-edited corrections, with approximately $315$ hours of audio and a $55$-hour subset fully transcribed. The resource supports tasks in SLA, SGEC, SGECF, and low-resource ASR for L2 English, includes disfluency handling, and provides guidance to prevent data leakage when used with LLMs. By enabling robust evaluation of spoken language assessment and feedback systems and facilitating research into automatic feedback and error-correction for spoken language, the corpus has potential to advance inclusive language-learning technologies and scalable assessment tools.

Abstract

We introduce the Speak & Improve Corpus 2025, a dataset of L2 learner English data with holistic scores and language error annotation, collected from open (spontaneous) speaking tests on the Speak & Improve learning platform. The aim of the corpus release is to address a major challenge to developing L2 spoken language processing systems, the lack of publicly available data with high-quality annotations. It is being made available for non-commercial use on the ELiT website. In designing this corpus we have sought to make it cover a wide-range of speaker attributes, from their L1 to their speaking ability, as well as providing manual annotations. This enables a range of language-learning tasks to be examined, such as assessing speaking proficiency or providing feedback on grammatical errors in a learner's speech. Additionally the data supports research into the underlying technology required for these tasks including automatic speech recognition (ASR) of low resource L2 learner English, disfluency detection or spoken grammatical error correction (GEC). The corpus consists of around 315 hours of L2 English learners audio with holistic scores, and a subset of audio annotated with transcriptions and error labels.

Figures (1)

• Figure 1: Annotation phases.