Zuva OCR API

openapi: 3.0.1
info:
  title: Zuva DocAI Classification OCR API
  description: The Zuva DocAI API is a RESTful contract and document AI API for extracting structured data from unstructured documents. It provides asynchronous file submission, OCR, field extraction across 1,400+ pre-built fields, multi-level document classification across 220+ document types, language detection, and a searchable fields catalog. All requests are authenticated with a Bearer API token and served from regional hosts (US and EU).
  termsOfService: https://zuva.ai/terms/zuva/
  contact:
    name: Zuva Support
    url: https://zuva.ai/
  version: '2.0'
servers:
- url: https://{region}.app.zuva.ai/api/v2
  description: Zuva DocAI regional API server
  variables:
    region:
      default: us
      description: Deployment region for the Zuva DocAI API.
      enum:
      - us
      - eu
security:
- bearerAuth: []
tags:
- name: OCR
  description: Optical character recognition.
paths:
  /ocr:
    post:
      operationId: createOcr
      tags:
      - OCR
      summary: Create an OCR request.
      requestBody:
        required: true
        content:
          application/json:
            schema:
              type: object
              required:
              - file_ids
              properties:
                file_ids:
                  type: array
                  items:
                    type: string
      responses:
        '202':
          description: OCR requests accepted and queued.
          content:
            application/json:
              schema:
                type: array
                items:
                  $ref: '#/components/schemas/OcrResult'
  /ocrs:
    get:
      operationId: getOcrs
      tags:
      - OCR
      summary: Query multiple OCR requests.
      parameters:
      - name: request_id
        in: query
        required: false
        description: Comma-separated list of OCR request IDs.
        schema:
          type: string
      responses:
        '200':
          description: OCR request statuses.
          content:
            application/json:
              schema:
                type: array
                items:
                  $ref: '#/components/schemas/OcrResult'
  /ocr/{request_id}:
    get:
      operationId: getOcr
      tags:
      - OCR
      summary: Retrieve an OCR request status.
      parameters:
      - $ref: '#/components/parameters/RequestId'
      responses:
        '200':
          description: OCR request status.
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/OcrResult'
  /ocr/{request_id}/text:
    get:
      operationId: getOcrText
      tags:
      - OCR
      summary: Retrieve OCR extracted text.
      parameters:
      - $ref: '#/components/parameters/RequestId'
      responses:
        '200':
          description: Extracted text.
          content:
            text/plain:
              schema:
                type: string
  /ocr/{request_id}/images:
    get:
      operationId: getOcrImages
      tags:
      - OCR
      summary: Download all page images as a zip.
      parameters:
      - $ref: '#/components/parameters/RequestId'
      responses:
        '200':
          description: Zip archive of page PNG images.
          content:
            application/zip:
              schema:
                type: string
                format: binary
  /ocr/{request_id}/images/{page_number}:
    get:
      operationId: getOcrImagePage
      tags:
      - OCR
      summary: Download a single page image.
      parameters:
      - $ref: '#/components/parameters/RequestId'
      - name: page_number
        in: path
        required: true
        schema:
          type: integer
      responses:
        '200':
          description: PNG image of the requested page.
          content:
            image/png:
              schema:
                type: string
                format: binary
  /ocr/{request_id}/layouts:
    get:
      operationId: getOcrLayouts
      tags:
      - OCR
      summary: Retrieve layout protobuf binary.
      parameters:
      - $ref: '#/components/parameters/RequestId'
      responses:
        '200':
          description: Layout protobuf binary.
          content:
            application/octet-stream:
              schema:
                type: string
                format: binary
components:
  schemas:
    OcrResult:
      type: object
      properties:
        request_id:
          type: string
        file_id:
          type: string
        status:
          type: string
        page_count:
          type: integer
        character_count:
          type: integer
        scan_quality:
          type: string
        scan_score:
          type: number
  parameters:
    RequestId:
      name: request_id
      in: path
      required: true
      description: The unique identifier of an asynchronous request.
      schema:
        type: string
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer
      description: 'Zuva API token passed in the Authorization header as `Authorization: Bearer <token>`.'