Overview
YomiToku is a proprietary document analysis engine specialized for Japanese. It integrates AI OCR plus layout and table parsing models, accurately structuring vertical text, multi column documents, and complex business forms. It supports a wide range of use cases, including generating data for RAG / search, creating searchable PDFs, and extracting information from table data.
We provide yomitoku-client as a client SDK to help you use this product more conveniently.For more details, please refer to the link below: https://github.com/MLism-Inc/yomitoku-client
For long-term or large-scale use, this product is also available through private offers.Please contact our support team for pricing information.
Highlights
- Each model is specifically trained for Japanese document images, supporting the recognition of over 7,000 Japanese characters, including vertical text and other layout structures unique to Japanese documents. (It also supports English documents.)
- Equipped with four AI models trained on Japanese datasets: text detection, text recognition, layout analysis, and table structure recognition. All models are independently trained and optimized for Japanese documents, delivering high-precision inference.
- By leveraging layout analysis, table structure parsing, and reading order estimation, it extracts information while preserving the semantic structure of the document layout.
Details
Features and programs
Financing for AWS Marketplace purchases
Pricing
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.g4dn.xlarge Inference (Batch) Recommended | Model inference on the ml.g4dn.xlarge instance type, batch mode | $10.00 |
ml.g4dn.xlarge Inference (Real-Time) Recommended | Model inference on the ml.g4dn.xlarge instance type, real-time mode | $10.00 |
ml.g5.xlarge Inference (Batch) | Model inference on the ml.g5.xlarge instance type, batch mode | $10.00 |
ml.g5.xlarge Inference (Real-Time) | Model inference on the ml.g5.xlarge instance type, real-time mode | $10.00 |
ml.g6.xlarge Inference (Real-Time) | Model inference on the ml.g6.xlarge instance type, real-time mode | $10.00 |
ml.g6.xlarge Inference (Batch) | Model inference on the ml.g6.xlarge instance type, batch mode | $10.00 |
ml.c7i.xlarge Inference (Batch) | Model inference on the ml.c7i.xlarge instance type, batch mode | $10.00 |
ml.c7i.2xlarge Inference (Batch) | Model inference on the ml.c7i.2xlarge instance type, batch mode | $10.00 |
ml.c7i.xlarge Inference (Real-Time) | Model inference on the ml.c7i.xlarge instance type, real-time mode | $10.00 |
ml.c7i.2xlarge Inference (Real-Time) | Model inference on the ml.c7i.2xlarge instance type, real-time mode | $10.00 |
Dimensions summary
Top-of-mind questions for buyers
Vendor refund policy
We do not support any refunds currently.
How can we make this page better?
Legal
Vendor terms and conditions
Content disclaimer
Delivery details
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Version release notes
Updated the inference engine to YomiToku Pro v1.8.0, which upgrades the text recognition model to v5. Recognition accuracy has improved for expressions specific to Japanese documents.
- Added support for composite glyphs such as "Japanese unit ligatures". They are expanded into their original katakana form in the output.
- Extended coverage of tate-chu-yoko (horizontal-in-vertical) text, including decimal numbers and day-of-week notation.
- Enclosed characters and unit symbols are now returned as they appear in the document. Unicode NFKC normalization has been removed so that the meaning of a character is preserved.
- Expanded coverage of emoji, symbols, and old-style / variant kanji through a larger character set and additional training data.
- More stable recognition of handwritten vertical text.
- Improved throughput on GPU instances (about 26% on ml.g6.xlarge and 14% on ml.g4dn.xlarge).
Additional details
Inputs
- Summary
Supported Content Types:
- application/pdf - PDF documents (multi-page supported)
- image/jpeg - JPEG images
- image/png - PNG images
- image/tiff - TIFF images
Request Body: Send the binary file data directly in the request body with appropriate Content-Type header.
Example Usage: aws sagemaker-runtime invoke-endpoint --endpoint-name --content-type "application/pdf" --body fileb://sample.pdf --region output.json
Sample Image Source: The sample images used in this documentation are from "Koho-shi November 2024" published by the Ministry of Internal Affairs and Communications of Japan.
- Input MIME type
- application/pdf, image/jpeg, image/png, image/tiff
Support
Vendor support
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.