43 lines
1.1 KiB
Python
43 lines
1.1 KiB
Python
import os, pymupdf, whisper, json
|
|
|
|
import cv2
|
|
import pytesseract
|
|
|
|
|
|
def read_files(path, output, filetypes=None):
|
|
if filetypes is None:
|
|
filetypes = ["pdf", "txt", "png", "jpg", "mp3"]
|
|
|
|
for root, dirs, files in os.walk(path):
|
|
for file in files:
|
|
if (file_type := file.split(".")[-1]) in filetypes:
|
|
output.append((file_type, os.path.join(path, file)))
|
|
|
|
for folder in dirs:
|
|
read_files(os.path.join(path, folder), output)
|
|
|
|
break
|
|
|
|
def extract_pdf_content(pdf_datei):
|
|
doc = pymupdf.open(pdf_datei)
|
|
a = ""
|
|
for page in doc:
|
|
a += page.get_text()
|
|
return a
|
|
|
|
|
|
def extract_mp3_content(mp3_datei):
|
|
model = whisper.load_model('tiny')
|
|
|
|
result = model.transcribe(str(mp3_datei), language='de', verbose=True)
|
|
|
|
# with open('transcript.json', "w") as f:
|
|
# json.dump(result['text'], f, indent=4)
|
|
|
|
return result["text"]
|
|
|
|
def extract_image_content(path):
|
|
pytesseract.pytesseract.tesseract_cmd = r'C:\Program Files\Tesseract-OCR\tesseract.exe'
|
|
img = cv2.imread(path)
|
|
return pytesseract.image_to_string(img)
|