---
id: amazon-textract-caller
version: "0.2.4"
license: Apache License Version 2.0
license_treatment: permissive
maintenance: active
---

# amazon-textract-caller — Amazon Textract Caller tools

License: permissive · Maintenance: active · Downloads: 911.7K/mo

## What it is and what it does

amazon-textract-caller is a Python wrapper that simplifies calling Amazon Textract, AWS's document analysis service. It abstracts away the complexity of choosing between synchronous and asynchronous APIs, handling file uploads to S3, and managing job polling. The package accepts documents from local filesystems or S3 buckets and returns structured JSON responses containing extracted text, tables, forms, and other document elements.

The wrapper supports Textract features like table detection, document queries, and custom adapters for specialized document types. It integrates with boto3 and botocore for AWS authentication and works alongside amazon-textract-response-parser for post-processing results. Most use cases involve a single call with a document path and optional feature flags, making it practical for rapid prototyping and production document-processing pipelines.

Use it for:

- Extract text and tables from scanned invoices or receipts stored in S3 for automated expense processing.
- Query multi-page documents to extract specific fields like employee SSN or gross pay from forms.
- Build a batch document ingestion pipeline that processes local PDFs and images without manual API management.
- Detect and extract structured data from forms with known layouts using table and form detection.
- Retrieve asynchronous job results with automatic polling instead of manual job-status checks.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Wrapper around Amazon Textract that handles document extraction from local files or S3, managing both synchronous and asynchronous API calls with support for table detection and document queries.

Yes. The package is actively maintained, has low install friction, carries a permissive license, and solves a real friction point in Textract integration. It is well-suited for teams already using AWS and needing to extract text or structured data from documents. No security vulnerabilities are known. Install if you are building document-processing workflows on AWS; skip if you do not use Textract or prefer direct boto3 calls.

## Install

pip install amazon-textract-caller
uv add amazon-textract-caller
poetry add amazon-textract-caller

## Installing amazon-textract-caller

Before you install: Low friction install with three runtime dependencies (boto3, botocore, amazon-textract-response-parser). Repository is actively maintained with recent commits and moderate GitHub presence (491 stars).

License in practice: Apache License Version 2.0 is permissive, allowing commercial use, modification, and distribution with minimal restrictions—suitable for most projects.

Quickstart:

pip install amazon-textract-caller

from amazon_textract_caller import call_textract

textract_json = call_textract(input_document="/path/to/file.png")

Requires AWS credentials configured for boto3 to authenticate with Amazon Textract service.

Verify before relying:

- Whether the package handles all document formats that Textract supports or has limitations on file types.
- Performance characteristics and latency for large documents or high-volume extraction workloads.
- Exact retry and error-handling behavior for failed Textract API calls.

## Package facts

- License: Apache License Version 2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 911.7K/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

amazon textract wrapper, document text extraction aws, ocr and table detection, textract async api caller, extract text from pdf images, aws textract helper library, document analysis textract, aws-integration, document-processing, ocr

[View on SkillFed](https://skillfed.io/packages/amazon-textract-caller) · [View on PyPI](https://pypi.org/project/amazon-textract-caller/)