Class OpenAIOcrCopilot

OpenAIOcrCopilot class

Provides OCR capabilities to extract text from PDF documents and images. The supported image types: PNG (.png), JPEG (.jpeg and .jpg), WEBP (.webp), non-animated GIF (.gif). Example usage of creating an OpenAI client, configuring options, and using the OCR copilot.

// Create AI client.
var openAiClient = OpenAIClient
    .CreateWithApiKey(ApiKey) // Create OpenAI client with the API key.
    //.WithOrganization("org_123") // Configure optional parameters.
    .Build(); // Build

// Create copilot options.
var options = OpenAIOcrCopilotOptions
    .Create() // Create options like this, or...
    //.Create(options => { options.Model = OpenAIModels.Gpt4O; }) // ...create using delegate.
    .WithDocument("DocumentInputPath"); // .WithDocument methods allows to add Document objects and paths to PDF documents and images.

// Create summary copilot.
IOcrCopilot ocrCopilot = AICopilotFactory.CreateOcrCopilot(openAiClient, options);

// Get text recognitions.
List<TextRecognitionResult> textRecognitions = await ocrCopilot.GetTextRecognitionResultAsync();

// Access to the extracted text.
string text = textRecognitions[0].OcrDetails[0].ExtractedText;
public class OpenAIOcrCopilot : IOcrCopilot

Constructors

NameDescription
OpenAIOcrCopilot(IOpenAIClient, IOcrCopilotOptions<OpenAIOcrCopilotOptions>)Initializes a new instance of the OpenAIOcrCopilot class.

Properties

NameDescription
HasContext { get; }

Methods

NameDescription
GetTextRecognitionResultAsync(CancellationToken?)

See Also