Update app.py

app.py

@@ -41,7 +41,6 @@ def load_whisper_model():
 
 def transcribe_audio(audio_file):
     model = load_whisper_model()
-    # Save temp file if not a path
     if isinstance(audio_file, str):
         audio_path = audio_file
     else:
@@ -53,6 +52,44 @@ def transcribe_audio(audio_file):
         os.remove(audio_path)
     return result["text"]
 
+def extract_text_from_file(uploaded_file):
+    # 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":
+        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']
+        if encoding:
+            content = content.decode(encoding, errors='ignore')
+        if file_type in ("html", "htm"):
+            soup = BeautifulSoup(content, "html.parser")
+            return soup.get_text()
+        elif file_type == "md":
+            html = markdown2.markdown(content)
+            soup = BeautifulSoup(html, "html.parser")
+            return soup.get_text()
+        elif file_type == "srt":
+            return re.sub(r"\d+\n\d{2}:\d{2}:\d{2},\d{3} --> .*?\n", "", content)
+        elif file_type in ("txt", "text"):
+            return content
+        else:
+            raise ValueError("Unsupported file type")
+
 def translate(text):
     translator = load_wolof_model()
     lang_tag = ">>wol<<"
@@ -80,34 +117,6 @@ def translate(text):
 
     return "\n".join(translated_output)
 
-def extract_text_from_file(uploaded_file):
-    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)
-        return "\n".join([para.text for para in doc.paragraphs])
-    else:
-        encoding = chardet.detect(content)['encoding']
-        if encoding:
-            content = content.decode(encoding, errors='ignore')
-        if file_type in ("html", "htm"):
-            soup = BeautifulSoup(content, "html.parser")
-            return soup.get_text()
-        elif file_type == "md":
-            html = markdown2.markdown(content)
-            soup = BeautifulSoup(html, "html.parser")
-            return soup.get_text()
-        elif file_type == "srt":
-            return re.sub(r"\d+\n\d{2}:\d{2}:\d{2},\d{3} --> .*?\n", "", content)
-        elif file_type in ("txt", "text"):
-            return content
-        else:
-            raise ValueError("Unsupported file type")
-
 def process_input(input_mode, text, audio_file, file_obj):
     input_text = ""
     if input_mode == "Text":
@@ -134,7 +143,7 @@ with gr.Blocks() as demo:
         input_mode = gr.Radio(choices=["Text", "Audio", "File"], label="Select input mode", value="Text")
 
     input_text = gr.Textbox(label="Enter English text", lines=10, visible=True)
-    audio_input = gr.Audio(label="Upload audio (.wav, .mp3, .m4a)", type="filepath")
+    audio_input = gr.Audio(label="Upload audio (.wav, .mp3, .m4a)", type="filepath", visible=False)
     file_input = gr.File(file_types=['.pdf', '.docx', '.html', '.htm', '.md', '.srt', '.txt'], label="Upload document", visible=False)
 
     extracted_text = gr.Textbox(label="Extracted / Transcribed Text", lines=10, interactive=False)