API Documentation
Documents
Document Conversion
Personal Earnings Statement
Certificate Of Origin
Dangerous Goods Declaration
International Consignment Note
Image To Text
Begin Image To Text Extraction
This endpoint processes a document using the image-to-text model. Note that the response can include an RFC 8288-style Link
header that may contain links to results. If multiple documents are sent for processing, there will be a link with rel=‘results’ for each document.
POST
/
v1-preview
/
models
/
image-to-text
using ABBYY.DocumentAI;
using ABBYY.DocumentAI.Schemas.Components;
var sdk = new DocumentAI(apiKeyAuth: "<YOUR_BEARER_TOKEN_HERE>");
BeginImageToTextTextExtractionRequestBody req = new BeginImageToTextTextExtractionRequestBody() {
InputSource = InputSource.CreateUrlInputSource(
new UrlInputSource() {
Url = "https://example.com/documents/invoice.png",
}
),
Options = new BeginImageToTextTextExtractionRequestBodyOptions() {
Languages = BeginImageToTextTextExtractionRequestBodyLanguages.En,
Handwriting = true,
},
};
var res = await sdk.Models.ImageToText.BeginTextExtractionAsync(req);
// handle response
[
{
"id": "wh23anb5xjf0ntw5taase5qz",
"name": "document.jpg",
"createdAt": "2025-02-07T17:44:06.949Z",
"model": "image-to-text",
"status": "Processed",
"pageCount": 10,
"errors": [
{
"message": "Total is required"
}
]
}
]
Authorizations
Our API offers authentication via API keys. You can obtain an API key from https://developer.abbyy.com
Body
application/json
Request to process a document using the image-to-text model
Response
201
application/json
A list containing the document(s) sent for processing
Was this page helpful?
using ABBYY.DocumentAI;
using ABBYY.DocumentAI.Schemas.Components;
var sdk = new DocumentAI(apiKeyAuth: "<YOUR_BEARER_TOKEN_HERE>");
BeginImageToTextTextExtractionRequestBody req = new BeginImageToTextTextExtractionRequestBody() {
InputSource = InputSource.CreateUrlInputSource(
new UrlInputSource() {
Url = "https://example.com/documents/invoice.png",
}
),
Options = new BeginImageToTextTextExtractionRequestBodyOptions() {
Languages = BeginImageToTextTextExtractionRequestBodyLanguages.En,
Handwriting = true,
},
};
var res = await sdk.Models.ImageToText.BeginTextExtractionAsync(req);
// handle response
[
{
"id": "wh23anb5xjf0ntw5taase5qz",
"name": "document.jpg",
"createdAt": "2025-02-07T17:44:06.949Z",
"model": "image-to-text",
"status": "Processed",
"pageCount": 10,
"errors": [
{
"message": "Total is required"
}
]
}
]
Assistant
Responses are generated using AI and may contain mistakes.