AmazonTextractConverter

Convert images and single-page PDFs to Haystack Documents using AWS Textract for OCR and structured data extraction.

Key Features

• Extracts text from images (JPEG, PNG, TIFF, BMP) and single-page PDFs using AWS Textract.
• Supports plain text OCR as well as structured data extraction (tables, forms, layouts).
• Supports natural language queries to extract specific information from documents.
• Automatically adds QUERIES feature type when queries are provided at runtime.
• Returns both Haystack Documents and raw Textract API responses.
• AWS credentials are resolved through AWS Secret parameters or the standard boto3 credential chain.

Configuration

1. Drag the AmazonTextractConverter component onto the canvas from the Component Library.
2. Click on the component to open the configuration panel.
3. On the General tab:
   1. Create secrets for your AWS credentials. Use AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, and AWS_DEFAULT_REGION as the environment variable names. For instructions, see Create Secrets.
   2. Optionally set feature_types to enable table extraction (TABLES), form extraction (FORMS), layout analysis (LAYOUT), or signature detection (SIGNATURES).
4. Go to the Advanced tab to configure store_full_path and boto3_config.

Connections

AmazonTextractConverter receives a list of file sources (paths or ByteStream objects). It outputs a list of Document objects containing the extracted text and a list of raw Textract API responses.

Source Code

To check this component's source code, open converter.py in the Haystack Core Integrations repository.

Usage Examples

Basic Configuration

  AmazonTextractConverter:
    type: haystack_integrations.components.converters.amazon_textract.AmazonTextractConverter
    init_parameters:
      aws_access_key_id:
        type: env_var
        env_vars:
        - AWS_ACCESS_KEY_ID
        strict: false
      aws_secret_access_key:
        type: env_var
        env_vars:
        - AWS_SECRET_ACCESS_KEY
        strict: false
      aws_region_name:
        type: env_var
        env_vars:
        - AWS_DEFAULT_REGION
        strict: false
      feature_types:

Using the Component in a Pipeline

# haystack-pipeline
components:
  AmazonTextractConverter:
    type: haystack_integrations.components.converters.amazon_textract.AmazonTextractConverter
    init_parameters:
      aws_access_key_id:
        type: env_var
        env_vars:
        - AWS_ACCESS_KEY_ID
        strict: false
      aws_secret_access_key:
        type: env_var
        env_vars:
        - AWS_SECRET_ACCESS_KEY
        strict: false
      aws_region_name:
        type: env_var
        env_vars:
        - AWS_DEFAULT_REGION
        strict: false
      feature_types:
      - TABLES
      - FORMS

  document_splitter:
    type: haystack.components.preprocessors.document_splitter.DocumentSplitter
    init_parameters:
      split_by: word
      split_length: 250
      split_overlap: 30

connections:
- sender: AmazonTextractConverter.documents
  receiver: document_splitter.documents

max_runs_per_component: 100

metadata: {}

Parameters

Inputs

Parameter	Type	Description
sources	List[Union[str, Path, ByteStream]]	A list of file paths, Path objects, or ByteStream objects to convert. Supported formats are JPEG, PNG, TIFF, BMP, and single-page PDF files (up to 10 MB).
meta	Optional[Union[Dict[str, Any], List[Dict[str, Any]]]]	Metadata to add to all documents, or a list of metadata dicts (one per source).
queries	Optional[List[str]]	A list of natural language queries to extract specific information from documents. Automatically enables the QUERIES Textract feature.

Outputs

Parameter	Type	Description
documents	List[Document]	The extracted text as Haystack Documents.
raw_textract_response	List[Dict]	The raw API responses from AWS Textract for each input file.

Init Parameters

These are the parameters you can configure in Pipeline Builder:

Parameter	Type	Default	Description
aws_access_key_id	Optional[Secret]	Secret.from_env_var("AWS_ACCESS_KEY_ID", strict=False)	The AWS access key ID.
aws_secret_access_key	Optional[Secret]	Secret.from_env_var("AWS_SECRET_ACCESS_KEY", strict=False)	The AWS secret access key.
aws_session_token	Optional[Secret]	Secret.from_env_var("AWS_SESSION_TOKEN", strict=False)	The AWS session token for temporary credentials.
aws_region_name	Optional[Secret]	Secret.from_env_var("AWS_DEFAULT_REGION", strict=False)	The AWS region to use (for example, us-east-1).
aws_profile_name	Optional[Secret]	Secret.from_env_var("AWS_PROFILE", strict=False)	The AWS named profile to use.
feature_types	Optional[List[str]]	None	Textract feature types to enable. Options are TABLES, FORMS, SIGNATURES, and LAYOUT. When not set, uses basic text detection only. The QUERIES type is added automatically when queries are provided at runtime.
store_full_path	bool	False	Whether to store the full file path in document metadata, or just the filename.
boto3_config	Optional[Dict[str, Any]]	None	A dictionary of additional configuration options for the boto3 Textract client.

Run Method Parameters

These are the parameters you can configure for the component's run() method. This means you can pass these parameters at query time through the API, in Playground, or when running a job. For details, see Modify Pipeline Parameters at Query Time.

Parameter	Type	Default	Description
sources	List[Union[str, Path, ByteStream]]		A list of file paths or streams to convert.
meta	Optional[Union[Dict, List[Dict]]]	None	Metadata to add to the resulting documents.
queries	Optional[List[str]]	None	Natural language queries to extract specific information.

Related Information

• Pipeline Components