Update app.py

app.py

@@ -233,92 +233,111 @@ def clone_voice_api(text_to_speak, reference_audio_url, exaggeration=0.6, cfg_pa
 
     temp_audio_path = None
     try:
-        print(f"API call received by clone_voice_api:")
-        print(f"  Text: {text_to_speak}")
-        print(f"  Audio URL type: {type(reference_audio_url)}")
-        print(f"  Audio URL preview: {str(reference_audio_url)[:100]}...")
-        print(f"  Parameters: exag={exaggeration}, cfg={cfg_pace}, seed={random_seed}, temp={temperature}")
-
-        if isinstance(reference_audio_url, str) and reference_audio_url.startswith('data:audio'):
+        print(f"=== API CALL DEBUG ===")
+        print(f"Text: {text_to_speak}")
+        print(f"Audio URL type: {type(reference_audio_url)}")
+        print(f"Audio URL length: {len(str(reference_audio_url)) if reference_audio_url else 0}")
+        print(f"Audio URL preview: {str(reference_audio_url)[:100]}...")
+        print(f"Parameters: exag={exaggeration}, cfg={cfg_pace}, seed={random_seed}, temp={temperature}")
+        
+        # Validate inputs
+        if not text_to_speak or text_to_speak.strip() == "":
+            return None, "Error: Please enter some text to speak."
+        
+        if not reference_audio_url:
+            return None, "Error: Please provide reference audio."
+            
+        print("Processing audio data...")
+        
+        if reference_audio_url.startswith('data:audio'):
             print("Processing base64 audio data...")
-            header, encoded = reference_audio_url.split(',', 1)
-            audio_data = base64.b64decode(encoded)
-            print(f"Decoded audio data size: {len(audio_data)} bytes")
-
-            if 'mp3' in header:
-                ext = '.mp3'
-            elif 'wav' in header:
-                ext = '.wav'
-            else:
-                ext = '.wav'
-
-            with tempfile.NamedTemporaryFile(delete=False, suffix=ext) as temp_file:
-                temp_file.write(audio_data)
-                temp_audio_path = temp_file.name
-
-            print(f"Created temporary audio file from base64: {temp_audio_path}")
-
-        elif isinstance(reference_audio_url, str) and reference_audio_url.startswith('http'):
+            try:
+                header, encoded = reference_audio_url.split(',', 1)
+                print(f"Header: {header}")
+                print(f"Encoded data length: {len(encoded)}")
+                
+                audio_data = base64.b64decode(encoded)
+                print(f"Decoded audio data size: {len(audio_data)} bytes")
+                
+                if 'mp3' in header:
+                    ext = '.mp3'
+                elif 'wav' in header:
+                    ext = '.wav'
+                else:
+                    ext = '.wav'
+                    
+                with tempfile.NamedTemporaryFile(delete=False, suffix=ext) as temp_file:
+                    temp_file.write(audio_data)
+                    temp_audio_path = temp_file.name
+                    
+                print(f"Created temporary audio file: {temp_audio_path}")
+                print(f"File exists: {os.path.exists(temp_audio_path)}")
+                print(f"File size: {os.path.getsize(temp_audio_path)} bytes")
+                
+            except Exception as audio_error:
+                print(f"Audio processing error: {audio_error}")
+                return None, f"Error processing audio data: {str(audio_error)}"
+                
+        elif reference_audio_url.startswith('http'):
             print("Processing HTTP audio URL...")
-            response = requests.get(reference_audio_url)
-            response.raise_for_status()
-            if reference_audio_url.endswith('.mp3'):
-                ext = '.mp3'
-            elif reference_audio_url.endswith('.wav'):
-                ext = '.wav'
-            else:
-                ext = '.wav' # Default
-            with tempfile.NamedTemporaryFile(delete=False, suffix=ext) as temp_file:
-                temp_file.write(response.content)
-                temp_audio_path = temp_file.name
-            print(f"Created temporary audio file from URL: {temp_audio_path}")
-        elif isinstance(reference_audio_url, str) and os.path.exists(reference_audio_url):
-             print("Using direct file path provided as string...")
-             temp_audio_path = reference_audio_url
+            try:
+                response = requests.get(reference_audio_url)
+                response.raise_for_status()
+                if reference_audio_url.endswith('.mp3'):
+                    ext = '.mp3'
+                elif reference_audio_url.endswith('.wav'):
+                    ext = '.wav'
+                else:
+                    ext = '.wav'
+                with tempfile.NamedTemporaryFile(delete=False, suffix=ext) as temp_file:
+                    temp_file.write(response.content)
+                    temp_audio_path = temp_file.name
+                print(f"Downloaded audio to: {temp_audio_path}")
+            except Exception as download_error:
+                print(f"Download error: {download_error}")
+                return None, f"Error downloading audio: {str(download_error)}"
         else:
-            # This case might occur if Gradio passes a TemporaryFileWrapper or similar
-            if hasattr(reference_audio_url, 'name'): # Check if it's a file-like object from Gradio
-                 temp_audio_path = reference_audio_url.name
-                 print(f"Using file path from Gradio object: {temp_audio_path}")
-            else:
-                print(f"Warning: Unrecognized audio input type or path: {reference_audio_url}. Assuming it's a direct path.")
-                temp_audio_path = str(reference_audio_url) # Fallback, attempt to use as path
-
-        if not temp_audio_path or not os.path.exists(temp_audio_path):
-            raise ValueError(f"Failed to obtain a valid audio file path from input: {reference_audio_url}")
-
-        print(f"Calling core clone_voice function with audio path: {temp_audio_path}")
+            print("Using direct file path...")
+            temp_audio_path = reference_audio_url
+
+        print(f"Calling clone_voice with:")
+        print(f"  Text: {text_to_speak}")
+        print(f"  Audio path: {temp_audio_path}")
+        print(f"  Parameters: {exaggeration}, {cfg_pace}, {random_seed}, {temperature}")
+        
+        # Call the main function
         audio_output, status = clone_voice(text_to_speak, temp_audio_path, exaggeration, cfg_pace, random_seed, temperature)
-        print(f"clone_voice returned: {type(audio_output)}, {status}")
+        
+        print(f"clone_voice returned:")
+        print(f"  Audio output type: {type(audio_output)}")
+        print(f"  Status: {status}")
 
-        # Clean up temporary file only if we created one from base64 or URL
-        if temp_audio_path and isinstance(reference_audio_url, str) and \
-           (reference_audio_url.startswith('data:audio') or reference_audio_url.startswith('http')):
+        # Cleanup
+        if temp_audio_path and temp_audio_path != reference_audio_url:
             try:
                 os.unlink(temp_audio_path)
                 print(f"Cleaned up temporary file: {temp_audio_path}")
-            except Exception as e:
-                print(f"Failed to clean up temp file {temp_audio_path}: {e}")
-
+            except Exception as cleanup_error:
+                print(f"Cleanup error: {cleanup_error}")
+                
         return audio_output, status
-
+        
     except Exception as e:
-        print(f"ERROR in clone_voice_api: {e}")
-        import traceback # Ensure traceback is imported here if not globally
+        print(f"=== CRITICAL ERROR ===")
+        print(f"Error type: {type(e)}")
+        print(f"Error message: {str(e)}")
+        import traceback
         traceback.print_exc()
-
-        # Attempt to clean up temporary file in case of error too
-        if temp_audio_path and isinstance(reference_audio_url, str) and \
-           (reference_audio_url.startswith('data:audio') or reference_audio_url.startswith('http')):
+        
+        # Cleanup on error
+        if temp_audio_path and temp_audio_path != reference_audio_url:
             try:
-                if os.path.exists(temp_audio_path): # Check existence before unlinking
-                    os.unlink(temp_audio_path)
-                    print(f"Cleaned up temporary file after error: {temp_audio_path}")
-            except Exception as e_clean:
-                 print(f"Failed to clean up temp file {temp_audio_path} after error: {e_clean}")
+                os.unlink(temp_audio_path)
+            except:
+                pass
+                
         return None, f"API Error: {str(e)}"
 
-
 def main():
     print("Starting Advanced Gradio interface...")