Convert a photo of a document into text

v1.0.0

Converts an uploaded photo of a document in common formats such as JPEG, PNG into text via Optical Character Recognition. This API is intended to be run on photos of documents, e.g. taken with a smartphone and supports cases where other content, such as a desk, are in the frame and the camera is crooked. If you want to OCR a scanned image, use the image/toText API call instead as it is designed for scanned images. Note: for free tier API keys, it is required to add a credit card to your account for security reasons, to use the free tier key with this API.

Input

{
  "type": "object",
  "properties": {
    "language": {
      "type": "string"
    },
    "recognitionMode": {
      "type": "string"
    }
  },
  "additionalProperties": false
}

Output

{
  "type": "object",
  "description": "Response from an OCR to text operation.  Includes the confience rating and converted text result.",
  "properties": {
    "MeanConfidenceLevel": {
      "type": "number",
      "format": "float",
      "description": "Confidence level rating of the OCR operation; ratings above 80% are strong."
    },
    "TextResult": {
      "type": "string",
      "description": "Converted text string from the image input."
    }
  }
}