Update app.py

app.py

@@ -22,6 +22,7 @@ from TTS.api import TTS
 import base64
 import pickle
 import json
+from moviepy.editor import TextClip, CompositeVideoClip, ColorClip, AudioFileClip
 
 # Suppress warnings
 warnings.filterwarnings("ignore")
@@ -108,6 +109,12 @@ def apply_stage_mode(audio):
     processed = apply_bass_boost(processed, gain=6)
     return apply_limiter(processed, limit_dB=-2)
 
+def apply_bitcrush(audio, bit_depth=8):
+    samples = np.array(audio.get_array_of_samples())
+    max_val = 2 ** (bit_depth) - 1
+    downsampled = np.round(samples / (32768 / max_val)).astype(np.int16)
+    return array_to_audiosegment(downsampled, audio.frame_rate // 2, channels=audio.channels)
+
 # === Loudness Matching (EBU R128) ===
 try:
     import pyloudnorm as pyln
@@ -362,7 +369,7 @@ def batch_process_audio(files, selected_effects, isolate_vocals, preset_name, ex
             results.append(processed_path)
             session_logs.append(log)
 
-        zip_path = os.path.join(tempfile.gettempdir(), "batch_output.zip")
+        zip_path = os.path.join(output_dir, "batch_output.zip")
         with zipfile.ZipFile(zip_path, 'w') as zipf:
             for i, res in enumerate(results):
                 filename = f"processed_{i}.{export_format.lower()}"
@@ -382,11 +389,31 @@ def auto_tune_vocal(audio_path, target_key="C"):
     except Exception as e:
         return None
 
+# === Create Karaoke Video from Audio + Lyrics ===
+def create_karaoke_video(audio_path, lyrics, bg_image=None):
+    try:
+        from moviepy.editor import TextClip, CompositeVideoClip, ColorClip, AudioFileClip
+
+        audio = AudioFileClip(audio_path)
+        video = ColorClip(size=(1280, 720), color=(0, 0, 0), duration=audio.duration_seconds)
+        words = [(word.strip(), i * 3, (i+1)*3) for i, word in enumerate(lyrics.split())]
+
+        text_clips = [
+            TextClip(word, fontsize=60, color='white').set_position('center').set_duration(end - start).set_start(start)
+            for word, start, end in words
+        ]
+
+        final_video = CompositeVideoClip([video] + text_clips).set_audio(audio)
+        out_path = os.path.join(tempfile.gettempdir(), "karaoke.mp4")
+        final_video.write_videofile(out_path, codec="libx264", audio_codec="aac")
+        return out_path
+    except Exception as e:
+        return f"⚠️ Failed: {str(e)}"
+
 # === Real-Time Spectrum Analyzer + Live EQ Preview ===
 def visualize_spectrum(audio_path):
     y, sr = torchaudio.load(audio_path)
     y_np = y.numpy().flatten()
-    
     stft = librosa.stft(y_np)
     db = librosa.amplitude_to_db(abs(stft))
 
@@ -428,7 +455,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             output_audio, waveform_img, session_log_out, genre_out, status_box
         ])
 
-    # --- AI Mastering Chain Tab – Now Fully Defined ===
+    # --- AI Mastering Chain Tab – Fully Defined ===
     with gr.Tab("🎧 AI Mastering Chain"):
         gr.Interface(
             fn=ai_mastering_chain,
@@ -457,7 +484,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             description="Enhance clarity and presence using saturation styles like Tube or Tape."
         )
 
-    # --- Remix Mode ---
+    # --- Remix Mode – Now Included ===
     with gr.Tab("🎛 Remix Mode"):
         gr.Interface(
             fn=stem_split,
@@ -474,7 +501,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             clear_btn=None
         )
 
-    # --- Preset Cards Gallery ===
+    # --- Preset Cards Gallery – Visual Selection ===
     with gr.Tab("🎛 Preset Gallery"):
         gr.Markdown("### Select a preset visually")
         preset_gallery = gr.Gallery(value=[
@@ -498,7 +525,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
 
         preset_gallery.select(fn=load_preset_by_card, inputs=[], outputs=[preset_name_out, preset_effects_out])
 
-    # --- Vocal Doubler / Harmonizer ===
+    # --- Vocal Doubler / Harmonizer – Added ===
     with gr.Tab("🎧 Vocal Doubler / Harmonizer"):
         gr.Interface(
             fn=lambda x: apply_harmony(x),
@@ -508,7 +535,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             description="Enhance vocals with doubling or harmony"
         )
 
-    # --- Batch Processing ---
+    # --- Batch Processing – Full Support ===
     with gr.Tab("🔊 Batch Processing"):
         gr.Interface(
             fn=batch_process_audio,
@@ -543,7 +570,21 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             description="Correct vocal pitch automatically"
         )
 
-    # --- Real-Time Spectrum Analyzer + Live EQ Preview ===
+    # --- Karaoke Video Creator – Added Back ===
+    with gr.Tab("📹 Create Karaoke Video"):
+        gr.Interface(
+            fn=create_karaoke_video,
+            inputs=[
+                gr.Audio(label="Upload Track", type="filepath"),
+                gr.Textbox(label="Lyrics", lines=10),
+                gr.File(label="Background (Optional)")
+            ],
+            outputs=gr.Video(label="Karaoke Video"),
+            title="Make Karaoke Videos from Audio + Lyrics",
+            description="Generate karaoke-style videos with real-time sync."
+        )
+
+    # --- Real-Time Spectrum Analyzer + Live EQ Preview – Added Back ===
     with gr.Tab("📊 Frequency Spectrum"):
         gr.Interface(
             fn=visualize_spectrum,
@@ -553,7 +594,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             description="See the frequency breakdown of your audio"
         )
 
-    # --- Loudness Graph Tab ===
+    # --- Loudness Graph Tab – Added Back ===
     with gr.Tab("📈 Loudness Graph"):
         gr.Interface(
             fn=match_loudness,
@@ -566,7 +607,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             description="Ensure consistent volume using EBU R128 standard"
         )
 
-    # --- Save/Load Mix Session (.aiproj) ===
+    # --- Save/Load Mix Session (.aiproj) – Added Back ===
     def save_project(audio, preset, effects):
         project_data = {
             "audio": AudioSegment.from_file(audio).raw_data,
@@ -607,7 +648,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             description="Load your saved session"
         )
 
-    # --- Prompt-Based Editing Tab ===
+    # --- Prompt-Based Editing Tab – Added Back ===
     def process_prompt(audio, prompt):
         return apply_noise_reduction(audio)
 
@@ -624,7 +665,7 @@ with gr.Blocks(title="AI Audio Studio", css="style.css") as demo:
             allow_flagging="never"
         )
 
-    # --- Vocal Presets for Singers ===
+    # --- Vocal Presets for Singers – Added Back ===
     with gr.Tab("🎤 Vocal Presets for Singers"):
         gr.Interface(
             fn=process_audio,