Layered visualisation of document, image and tabular datasets

Products

Ready-to-train datasets, licensed and documented.

Each catalogue dataset ships with a datasheet, a licence covering commercial model training, a held-out evaluation split and a provenance record for every source.

Catalogue

Available now

Sample splits of 500 records are available for any dataset on request, at no cost.

BX-DOC-01Documents

Global Invoice OCR Corpus

Invoices and receipts from 38 countries with 24 extracted fields per page, line-item tables and full bounding-box coordinates.

Volume
1.2M pages
Formats
JSONL + PNG
BX-VIS-04Images

Retail Shelf Vision Set

Store-shelf photography with SKU-level boxes, facing counts, price-tag regions and out-of-stock gap annotations.

Volume
480K images
Formats
COCO / YOLO
BX-SPX-02Audio

Multilingual Call Speech

Consented contact-centre audio in 21 languages with verbatim transcripts, diarisation and accent metadata.

Volume
9,400 hours
Formats
WAV + CTM
BX-NLP-07Text

Instruction Tuning Pack

Human-authored instruction and response pairs across reasoning, coding, extraction and refusal categories, with rubric scores.

Volume
620K pairs
Formats
JSONL
BX-AUT-03Video + LiDAR

Urban Driving Perception

Synchronised camera and LiDAR sequences with 3D cuboids, tracking IDs, lane geometry and weather condition tags.

Volume
310 hours
Formats
nuScenes-style
BX-MED-05Documents

Clinical Form Extraction

De-identified clinical intake and claims forms with structured field extraction, reviewed by medical coders.

Volume
180K forms
Formats
JSONL

Custom builds

Nothing in the catalogue fits?

Collection

We recruit consented contributors and capture net-new data to your protocol — imagery, speech, handwriting or behavioural.

Synthesis & augmentation

Programmatic generation and augmentation to fill class imbalance, with human review on every synthetic sample kept.

Licensing & compliance

Provenance tracking, consent records, PII scrubbing and a datasheet you can hand to your legal team.

Need labelled data by next sprint?

Send us a spec, a schema, or a handful of raw files. We return a pilot batch of 1,000 records with a quality report within five working days — free of charge.

info@betalen.in