Update app.py

app.py

@@ -23,6 +23,7 @@ from faster_whisper import WhisperModel
 from mutagen.mp3 import MP3
 from mutagen.id3 import ID3, TIT2, TPE1, TALB, TYER
 from TTS.api import TTS
+import pickle
 
 # Suppress warnings
 warnings.filterwarnings("ignore")
@@ -318,6 +319,11 @@ def save_project(audio_path, preset_name, effects):
         pickle.dump(project_data, f)
     return out_path
 
+def load_project(project_file):
+    with open(project_file.name, "rb") as f:
+        data = pickle.load(f)
+    return data["preset"], data["effects"]
+
 # === Trim Silence Automatically (VAD) ===
 def detect_silence(audio_file, silence_threshold=-50.0, min_silence_len=1000):
     audio = AudioSegment.from_file(audio_file)
@@ -345,25 +351,12 @@ def mix_tracks(track1, track2, volume_offset=0):
     mixed.export(out_path, format="wav")
     return out_path
 
-# === Load Custom Plugins ===
-def load_plugin(plugin_file, audio_file):
-    try:
-        from importlib.util import spec_from_file_location, module_from_spec
-        spec = spec_from_file_location("plugin", plugin_file.name)
-        plugin = module_from_spec(spec)
-        spec.loader.exec_module(plugin)
-        
-        # Run plugin
-        audio = AudioSegment.from_file(audio_file)
-        processed = plugin.process(audio)
-        
-        out_path = os.path.join(tempfile.gettempdir(), "plugin_output.wav")
-        processed.export(out_path, format="wav")
-        return out_path
-    except Exception as e:
-        return f"⚠️ Plugin error: {str(e)}"
+# === Voice Style Transfer (Dummy) ===
+def apply_style_transfer(audio_path, mood="Happy"):
+    # Replace with real model later
+    return audio_path
 
-# === UI ===
+# === UI Setup ===
 effect_options = [
     "Noise Reduction",
     "Compress Dynamic Range",
@@ -443,13 +436,13 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             clear_btn=None
         )
 
-    # --- Transcribe & Edit ---
+    # --- Transcribe & Edit Tab ===
     with gr.Tab("📝 Transcribe & Edit"):
         gr.Interface(
             fn=transcribe_audio,
             inputs=gr.Audio(label="Upload Audio", type="filepath"),
             outputs=gr.Textbox(label="Transcribed Text", lines=10),
-            title="Transcribe & Edit Spoken Content",
+            title="Transcribe Spoken Content",
             description="Convert voice to text and edit it before exporting again."
         )
 
@@ -479,28 +472,62 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             description="Detect speaker similarity using AI."
         )
 
-    # --- Auto-Save Sessions ===
-    def encode_preset(selected_effects, preset_name, export_format):
-        import base64
-        import json
-        data = {"effects": selected_effects, "preset": preset_name, "format": export_format}
-        encoded = base64.b64encode(json.dumps(data).encode()).decode()
-        return f"https://your-space-url?preset={encoded}"
+    # --- Voice Cloning (Dummy) ===
+    def clone_voice(*args):
+        return "⚠️ Voice cloning requires additional setup"
 
-    with gr.Tab("🧾 Share Session"):
+    with gr.Tab("🎭 Voice Cloning (Dubbing)"):
         gr.Interface(
-            fn=encode_preset,
+            fn=clone_voice,
             inputs=[
-                gr.CheckboxGroup(choices=effect_options, label="Effects"),
-                gr.Dropdown(choices=preset_names, label="Preset"),
-                gr.Dropdown(choices=["MP3", "WAV"], label="Format")
+                gr.File(label="Source Voice Clip"),
+                gr.File(label="Target Voice Clip"),
+                gr.Textbox(label="Text to Clone", lines=5)
             ],
-            outputs=gr.Textbox(label="Shareable Link", lines=1),
-            title="Save Your Settings and Share Them",
-            description="Generate a link to share your effect chain with others."
+            outputs=gr.Audio(label="Cloned Output", type="filepath"),
+            title="Replace One Voice With Another",
+            description="Clone voice from source to target speaker using AI"
         )
 
-    # --- VAD – Detect & Remove Silence === 
+    # --- Auto-Save / Resume Sessions ===
+    session_state = gr.State()
+
+    def auto_save_session(audio, preset, effects):
+        return {"audio": audio, "preset": preset, "effects": effects}
+
+    def auto_load_session(session):
+        if session and "audio" in session:
+            return session["audio"], session["preset"], session["effects"]
+        return None, None, None
+
+    with gr.Tab("🧾 Auto-Save & Resume"):
+        gr.Interface(
+            fn=auto_save_session,
+            inputs=[
+                gr.Audio(label="Upload Audio", type="filepath"),
+                gr.Dropdown(choices=preset_names, label="Used Preset", value=preset_names[0]),
+                gr.CheckboxGroup(choices=effect_options, label="Applied Effects")
+            ],
+            outputs=session_state,
+            title="Auto-Save Your Session",
+            description="Save your current state and resume editing later",
+            allow_flagging="never"
+        )
+
+        gr.Interface(
+            fn=auto_load_session,
+            inputs=session_state,
+            outputs=[
+                gr.Audio(label="Loaded Audio", type="filepath"),
+                gr.Dropdown(choices=preset_names, label="Loaded Preset"),
+                gr.CheckboxGroup(choices=effect_options, label="Loaded Effects")
+            ],
+            title="Resume Last Session",
+            description="Reload your last edit state",
+            allow_flagging="never"
+        )
+
+    # --- VAD – Detect & Remove Silence ===
     with gr.Tab("✂️ Trim Silence Automatically"):
         gr.Interface(
             fn=detect_silence,
@@ -511,10 +538,10 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             ],
             outputs=gr.File(label="Trimmed Output"),
             title="Auto-Detect & Remove Silence",
-            description="Trim intros/outs or between speech automatically"
+            description="Detect and trim silence at start/end or between words"
         )
 
-    # --- Load/Save Project ===
+    # --- Load/Save Project File (.aiproj) ===
     with gr.Tab("📁 Save/Load Project"):
         gr.Interface(
             fn=save_project,
@@ -528,6 +555,17 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             description="Save your session, effects, and settings in one file to reuse later."
         )
 
+        gr.Interface(
+            fn=load_project,
+            inputs=gr.File(label="Upload .aiproj File"),
+            outputs=[
+                gr.Dropdown(choices=preset_names, label="Loaded Preset"),
+                gr.CheckboxGroup(choices=effect_options, label="Loaded Effects")
+            ],
+            title="Resume Last Project",
+            description="Load your saved session"
+        )
+
     # --- Mix Two Tracks ===
     with gr.Tab("🔀 Mix Two Tracks"):
         gr.Interface(
@@ -542,17 +580,17 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             description="Mix or subtract two audio files."
         )
 
-    # --- Custom Effect Plugin System ===
-    with gr.Tab("🧩 Load Custom Effect"):
+    # --- Voice Style Transfer (Dummy) ===
+    with gr.Tab("🧠 Voice Style Transfer"):
         gr.Interface(
-            fn=load_plugin,
+            fn=apply_style_transfer,
             inputs=[
-                gr.File(label="Upload .py plugin"),
-                gr.Audio(label="Upload Audio", type="filepath")
+                gr.Audio(label="Upload Voice Clip", type="filepath"),
+                gr.Radio(["Happy", "Sad", "Angry", "Calm"], label="Choose Tone")
             ],
-            outputs=gr.Audio(label="Processed Output", type="filepath"),
-            title="Run Your Own Python Effect",
-            description="Upload a .py file with a 'process' function"
+            outputs=gr.Audio(label="Stylized Output", type="filepath"),
+            title="Change Emotional Tone of Voice",
+            description="Shift the emotional style of any voice clip."
         )
 
 demo.launch()