CEFR-Graded English Teaching Corpus (A1–C2) with Native British Audio
Audio, Speech & Acoustic Datasets
Tags and Keywords
$16,500
About
A complete, teacher-authored English language teaching corpus covering every CEFR level from A1 to C2. It contains 600 lessons (100 per level, organised into 10 courses per level) with aligned native British English audio, plus an overview for each level and a high-level overview of the full corpus.
The entire corpus was written and recorded by a single author for The English Classroom, a teaching platform used by 7,000+ registered teachers in 80+ countries, including a UNRWA deployment. Each lesson is mapped to both a CEFR level and the UN Sustainable Development Goals, so the content is consistently levelled and tied to themes.
The rights position is clean: one author, no third-party material, and all rights held by The English Classroom AB.
Data Product Features
- CEFR level: every lesson is tagged A1, A2, B1, B2, C1 or C2
- Course structure: 10 courses per level, each with sequenced lessons
- SDG theme: each lesson is mapped to one of the UN Sustainable Development Goals, which is unique to this corpus
- Lesson slides: instructional content written for the stated level
- Teacher notes: a separate document for each lesson, keyed to slide numbers, in the same format across all 600 lessons
- Audio: 2,500+ native British English clips, aligned to individual slides
- Activity types: a range of classroom activities within each lesson
- Level overviews: one overview document per CEFR level (6 in total) summarising courses, lessons and progression
- Corpus overview: a high-level document describing the full corpus structure, levels and themes
Distribution
- Format: lesson PDFs, SCORM packages (slides with per-slide audio), teacher-note PDFs, and overview documents
- Data Volume: 600 lessons, 60 courses, 6 CEFR levels, 2,500+ audio clips, 6 level overviews and 1 corpus overview, approx. 2.2 GB
- Audio note: the audio is compressed as delivered within the lesson packages, and no uncompressed master recordings exist
- Delivery: custom delivery after licence agreement; a sample is available on request
Usage
This data product is ideal for a variety of applications:
- LLM fine-tuning for language education: training models to produce or assess English at a specific CEFR level
- Proficiency classification: labelled examples for automatic CEFR level detection
- Educational content generation: reference material for AI tutoring and lesson-authoring tools
- Text-audio alignment: pairing lesson content with native-speaker audio for listening comprehension tasks and ASR evaluation
Coverage
- Geographic Coverage: Global. The content is in British English and is used in teaching in 80+ countries.
- Time Range: created 2019–2020
- Demographics: written for learners of English as a second or foreign language at every level from beginner to proficient. All audio is by one adult native British English speaker.
License
Non-exclusive commercial licence. Contact us for terms and pricing. All rights are held by The English Classroom AB.
AI Training Rights
Licensee is granted a non-exclusive, worldwide right, for the term agreed in the licence, to:
- Use the Data Product to train, fine-tune and evaluate machine learning models, including large language models.
- Incorporate Data Product content into models and commercialise resulting model outputs.
- Create derivative works (model weights, embeddings, etc.) for any lawful purpose not excluded below.
Restrictions:
- The audio may not be used for voice cloning, text-to-speech, speech synthesis, or training any model to imitate the speaker's voice.
- The Data Product itself may not be sold, redistributed or shared outside of licensed usage.
- Models may not be used to reproduce substantial portions of the lessons verbatim.
- No exclusive rights are granted.
- Licensee must comply with all applicable laws, including data protection and privacy regulations.
Who Can Use It
- AI labs and model developers: fine-tuning and evaluating models for language education and CEFR-aware generation.
- EdTech companies: building AI tutors, automated lesson authoring, and level-assessment tools.
- Researchers: applied linguistics, language assessment and second language acquisition studies.
- Speech and ASR teams: evaluating recognition and comprehension against levelled native-speaker audio (synthesis and voice cloning excluded).
Data Dictionary
Each lesson is one record. Key fields and components:
| Column Name | Data Type | Description | Possible Values/Notes |
|---|---|---|---|
| cefr_level | String | CEFR proficiency level of the lesson | A1, A2, B1, B2, C1, C2 |
| course_title | String | Course the lesson belongs to | 10 courses per level |
| lesson_number | Integer | Position of the lesson within its course | Sequential |
| lesson_title | String | Title of the lesson | Free text |
| sdg_theme | String | UN Sustainable Development Goal the lesson maps to | SDG 1–17 |
| slides | Document | Lesson slide content | PDF and SCORM |
| teacher_notes | Document | Presenter notes keyed to slide numbers | PDF, consistent format |
| audio | Audio | Native British English clips aligned to slides | Compressed, within SCORM packages |
| activity_type | String | Classroom activity used on a slide or section | Varies by lesson |
| level_overview | Document | Summary of courses, lessons and progression for one CEFR level | 6 documents, one per level |
| corpus_overview | Document | High-level description of the full corpus | 1 document |
All content was written and recorded by one person, so style, levelling and teacher-note format are consistent across the full corpus. Buyers can review the corpus overview and level overviews before licensing, and a sample lesson from each level is available on request.
TRUSTLoading...
$16,500
Download Dataset in Unknown Format
Recommended Datasets
Loading recommendations...