Update app.py

app.py

@@ -43,9 +43,18 @@ def load_models():
 def load_whisper_model():
     return whisper.load_model("base")
 
-def transcribe_audio(audio_path):
-    whisper_model = load_whisper_model()
-    return whisper_model.transcribe(audio_path)["text"]
+def transcribe_audio(audio_file):
+    model = load_whisper_model()
+    if isinstance(audio_file, str):
+        audio_path = audio_file
+    else:
+        with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as tmp:
+            tmp.write(audio_file.read())
+            audio_path = tmp.name
+    result = model.transcribe(audio_path)
+    if not isinstance(audio_file, str):
+        os.remove(audio_path)
+    return result["text"]
 
 def translate(text, target_lang):
     en_dar_translator, en_hau_translator, en_wol_translator = load_models()
@@ -90,14 +99,24 @@ def translate(text, target_lang):
 
 # --- Extract text from file ---
 def extract_text_from_file(uploaded_file):
-    file_type = uploaded_file.name.split('.')[-1].lower()
-    content = uploaded_file.read()
+    # Handle both filepath (str) and file-like object
+    if isinstance(uploaded_file, str):
+        file_path = uploaded_file
+        file_type = file_path.split('.')[-1].lower()
+        with open(file_path, "rb") as f:
+            content = f.read()
+    else:
+        file_type = uploaded_file.name.split('.')[-1].lower()
+        content = uploaded_file.read()
 
     if file_type == "pdf":
         with fitz.open(stream=content, filetype="pdf") as doc:
             return "\n".join([page.get_text() for page in doc])
     elif file_type == "docx":
-        doc = docx.Document(uploaded_file)
+        if isinstance(uploaded_file, str):
+            doc = docx.Document(file_path)
+        else:
+            doc = docx.Document(uploaded_file)
         return "\n".join([para.text for para in doc.paragraphs])
     else:
         encoding = chardet.detect(content)['encoding']