14 lines
354 B
Markdown
14 lines
354 B
Markdown
|
```elixir title="Elixir"
|
||
|
|
alias Kreuzberg.ExtractionConfig
|
||
|
|
|
||
|
|
config = %ExtractionConfig{
|
||
|
|
ocr: %{"enabled" => true, "backend" => "paddle-ocr", "language" => "en"}
|
||
|
|
# Add "model_tier" => "server" for max accuracy
|
||
|
|
}
|
||
|
|
|
||
|
|
{:ok, result} = Kreuzberg.extract_file("scanned_document.pdf", nil, config)
|
||
|
|
|
||
|
|
IO.puts("OCR Extracted content:")
|
||
|
|
IO.puts(result.content)
|
||
|
|
```
|