Notes

import fitz
import requests

def translate_text(text, api_key, endpoint, target_language):
url = f"{endpoint}/translate?api-version=3.0"
headers = {
"Ocp-Apim-Subscription-Key": api_key,
"Content-Type": "application/json",
}
body = [{"text": text}]
params = {"to": target_language}

response = requests.post(url, headers=headers, params=params, json=body)
if response.status_code == 200:
translation = response.json()[0]["translations"][0]["text"]
return translation
else:
print("Error:", response.status_code, response.text)
return None

def pdf_to_text(pdf_path):
# Open the PDF file
doc = fitz.open(pdf_path)

# Variable to store the final text
full_text = ""

# Go through each page
for page in doc:
print(f"Extracting text from page {page.number}...")
# Extract text from the page
text = page.get_text()
# Add this text to the full text
full_text += text

# Close the PDF document
doc.close()

return full_text

if __name__ == "__main__":
# PDF file path
pdf_path = "my_path.pdf"

# Translate API credentials
api_key = "your_api_key"
endpoint = "your_endpoint"

# Extract text from PDF
text = pdf_to_text(pdf_path)

# Translate extracted text to Spanish
translated_text_spanish = translate_text(text, api_key, endpoint, "es")
print("Translated text (Spanish):", translated_text_spanish)

# Translate extracted text to French
translated_text_french = translate_text(text, api_key, endpoint, "fr")
print("Translated text (French):", translated_text_french)