puuush

puzshsahc usah

lib.py

@@ -1,4 +1,4 @@
-import os, pymupdf
+import os, pymupdf, whisper, json
 
 def read_files(path, output, filetypes=None):
     if filetypes is None:
@@ -19,4 +19,13 @@ def extract_pdf_content(pdf_datei):
     a = ""
     for page in doc:
         a += page.get_text()
-    return a
+    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)

main.py

@@ -9,7 +9,7 @@ def retrieve_file_contents(path):
         "jpg": lambda path: "",
         "png": lambda path: "",
         "txt": lambda path: lib.extract_pdf_content(path),
-        "mp3": lambda path: "test mp3"
+        "mp3": lambda path: lib.extract_mp3_content(path),
     }
 
     lib.read_files(path, files := [])