Delete app.py

app.py

@@ -1,451 +0,0 @@
-#!/usr/bin/env python3
-"""
-Madverse Music: AI Music Detection Web App
-A Streamlit interface for testing AI-generated music detection
-"""
-
-import streamlit as st
-import torch
-import librosa
-import numpy as np
-import tempfile
-import os
-from pathlib import Path
-import time
-
-# Configure the page
-st.set_page_config(
-    page_title="Madverse Music: AI Music Detector",
-    page_icon="🎵",
-    layout="wide",
-    initial_sidebar_state="expanded"
-)
-
-# Custom CSS for styling
-st.markdown("""
-<style>
-    .main-header {
-        font-size: 3rem;
-        color: #1f77b4;
-        text-align: center;
-        margin-bottom: 2rem;
-        background: linear-gradient(90deg, #667eea 0%, #764ba2 100%);
-        -webkit-background-clip: text;
-        -webkit-text-fill-color: transparent;
-        font-weight: bold;
-    }
-    
-    .sub-header {
-        font-size: 1.2rem;
-        color: #666;
-        text-align: center;
-        margin-bottom: 3rem;
-    }
-    
-    .result-box {
-        padding: 1.5rem;
-        border-radius: 10px;
-        margin: 1rem 0;
-        border-left: 5px solid;
-    }
-    
-    .real-music {
-        background-color: #d4edda;
-        border-left-color: #28a745;
-        color: #155724;
-    }
-    
-    .fake-music {
-        background-color: #f8d7da;
-        border-left-color: #dc3545;
-        color: #721c24;
-    }
-    
-    .info-box {
-        background-color: #e3f2fd;
-        padding: 1rem;
-        border-radius: 8px;
-        border-left: 4px solid #2196f3;
-        margin: 1rem 0;
-    }
-    
-    .metric-card {
-        background: white;
-        padding: 1rem;
-        border-radius: 8px;
-        box-shadow: 0 2px 4px rgba(0,0,0,0.1);
-        text-align: center;
-        margin: 0.5rem;
-    }
-</style>
-""", unsafe_allow_html=True)
-
-# Initialize session state
-if 'model' not in st.session_state:
-    st.session_state.model = None
-if 'model_loaded' not in st.session_state:
-    st.session_state.model_loaded = False
-
-@st.cache_resource
-def load_model():
-    """Load the AI model (cached for performance)"""
-    try:
-        from sonics import HFAudioClassifier
-        model = HFAudioClassifier.from_pretrained("awsaf49/sonics-spectttra-alpha-120s")
-        model.eval()
-        return model
-    except Exception as e:
-        st.error(f"Failed to load model: {e}")
-        return None
-
-def classify_audio_file(audio_file, model):
-    """Classify uploaded audio file"""
-    try:
-        # Create temporary file
-        with tempfile.NamedTemporaryFile(delete=False, suffix='.wav') as tmp_file:
-            tmp_file.write(audio_file.getvalue())
-            tmp_file_path = tmp_file.name
-        
-        # Load audio (model uses 16kHz sample rate)
-        audio, sr = librosa.load(tmp_file_path, sr=16000)
-        
-        # Convert to tensor and add batch dimension
-        audio_tensor = torch.FloatTensor(audio).unsqueeze(0)
-        
-        # Get prediction
-        with torch.no_grad():
-            output = model(audio_tensor)
-            
-            # Convert logit to probability using sigmoid
-            prob = torch.sigmoid(output).item()
-            
-            # Classify: prob < 0.5 = Real, prob >= 0.5 = Fake
-            if prob < 0.5:
-                classification = "Real"
-                confidence = (1 - prob) * 2  # Convert to 0-1 scale
-            else:
-                classification = "Fake"
-                confidence = (prob - 0.5) * 2  # Convert to 0-1 scale
-        
-        # Cleanup
-        os.unlink(tmp_file_path)
-        
-        return {
-            "classification": classification,
-            "confidence": min(confidence, 1.0),  # Cap at 1.0
-            "probability": prob,
-            "raw_score": output.item(),
-            "duration": len(audio) / sr
-        }
-        
-    except Exception as e:
-        # Cleanup on error
-        if 'tmp_file_path' in locals() and os.path.exists(tmp_file_path):
-            os.unlink(tmp_file_path)
-        raise e
-
-# Main UI
-def main():
-    # Header
-    st.markdown('<h1 class="main-header">🎵 Madverse Music AI Detector</h1>', unsafe_allow_html=True)
-    st.markdown('<p class="sub-header">Upload audio files to detect if they\'re AI-generated or human-created</p>', unsafe_allow_html=True)
-    
-    # Sidebar
-    with st.sidebar:
-        st.header("🔬 Model Information")
-        
-        # Model loading status
-        if not st.session_state.model_loaded:
-            if st.button("🚀 Load AI Model", type="primary"):
-                with st.spinner("Loading Madverse Music AI model..."):
-                    st.session_state.model = load_model()
-                    if st.session_state.model:
-                        st.session_state.model_loaded = True
-                        st.success("✅ Model loaded successfully!")
-                        st.rerun()
-                    else:
-                        st.error("❌ Failed to load model")
-        else:
-            st.success("✅ Model Ready")
-            
-            # Model specs
-            st.markdown("""
-            **Model Details:**
-            - **Architecture**: SpecTTTra-α
-            - **Duration**: 120 seconds
-            - **Sample Rate**: 16kHz
-            - **F1 Score**: 97%
-            - **Sensitivity**: 96%
-            - **Specificity**: 99%
-            """)
-            
-            if st.button("🔄 Reload Model"):
-                st.session_state.model_loaded = False
-                st.session_state.model = None
-                st.cache_resource.clear()
-                st.rerun()
-    
-    # Main content
-    if st.session_state.model_loaded:
-        st.markdown("### 📁 Upload Audio File")
-        
-        # File uploader with bulk support
-        uploaded_files = st.file_uploader(
-            "Choose audio file(s)",
-            type=['wav', 'mp3', 'flac', 'm4a', 'ogg'],
-            accept_multiple_files=True,
-            help="Supported formats: WAV, MP3, FLAC, M4A, OGG (Max 200MB each). You can upload multiple files for bulk analysis."
-        )
-        
-        if uploaded_files:
-            # Display summary info for multiple files
-            if len(uploaded_files) == 1:
-                # Single file - show detailed info
-                uploaded_file = uploaded_files[0]
-                col1, col2, col3 = st.columns(3)
-                with col1:
-                    st.metric("📝 Filename", uploaded_file.name)
-                with col2:
-                    file_size_mb = uploaded_file.size / (1024 * 1024)
-                    st.metric("📊 File Size", f"{file_size_mb:.2f} MB")
-                with col3:
-                    st.metric("🎵 Format", uploaded_file.type)
-                
-                # Audio player
-                st.audio(uploaded_file, format='audio/wav')
-            else:
-                # Multiple files - show summary
-                total_size = sum(f.size for f in uploaded_files) / (1024 * 1024)
-                col1, col2, col3 = st.columns(3)
-                with col1:
-                    st.metric("📝 Files Selected", len(uploaded_files))
-                with col2:
-                    st.metric("📊 Total Size", f"{total_size:.2f} MB")
-                with col3:
-                    formats = list(set(f.type.split('/')[-1].upper() for f in uploaded_files))
-                    st.metric("🎵 Formats", ", ".join(formats))
-                
-                # Show file list
-                with st.expander(f"📋 File List ({len(uploaded_files)} files)"):
-                    for i, file in enumerate(uploaded_files, 1):
-                        size_mb = file.size / (1024 * 1024)
-                        st.write(f"{i}. **{file.name}** ({size_mb:.2f} MB)")
-            
-            # Analyze button
-            if len(uploaded_files) == 1:
-                button_text = "🔍 Analyze Audio"
-            else:
-                button_text = f"🔍 Analyze {len(uploaded_files)} Files"
-            
-            if st.button(button_text, type="primary", use_container_width=True):
-                try:
-                    if len(uploaded_files) == 1:
-                        # Single file analysis
-                        with st.spinner("🧠 Analyzing audio with AI..."):
-                            start_time = time.time()
-                            result = classify_audio_file(uploaded_files[0], st.session_state.model)
-                            processing_time = time.time() - start_time
-                        
-                        # Display results for single file
-                        st.markdown("### 🎯 Analysis Results")
-                        
-                        # Main result
-                        if result["classification"] == "Real":
-                            st.markdown(f"""
-                            <div class="result-box real-music">
-                                <h3>🎤 Human-Created Music</h3>
-                                <p>This audio appears to be created by human artists.</p>
-                            </div>
-                            """, unsafe_allow_html=True)
-                        else:
-                            st.markdown(f"""
-                            <div class="result-box fake-music">
-                                <h3>🤖 AI-Generated Music</h3>
-                                <p>This audio appears to be generated by AI (like Suno, Udio, etc.)</p>
-                            </div>
-                            """, unsafe_allow_html=True)
-                        
-                        # Detailed metrics for single file
-                        col1, col2, col3, col4 = st.columns(4)
-                        
-                        with col1:
-                            st.metric(
-                                "🎯 Classification",
-                                result["classification"],
-                                help="AI model's prediction"
-                            )
-                        
-                        with col2:
-                            confidence_pct = result["confidence"] * 100
-                            st.metric(
-                                "📊 Confidence",
-                                f"{confidence_pct:.1f}%",
-                                help="How confident the model is in its prediction"
-                            )
-                        
-                        with col3:
-                            st.metric(
-                                "⏱️ Duration",
-                                f"{result['duration']:.1f}s",
-                                help="Audio file duration"
-                            )
-                        
-                        with col4:
-                            st.metric(
-                                "⚡ Processing Time",
-                                f"{processing_time:.2f}s",
-                                help="Time taken to analyze"
-                            )
-                        
-                        # Technical details (expandable)
-                        with st.expander("🔬 Technical Details"):
-                            col1, col2 = st.columns(2)
-                            with col1:
-                                st.metric("Raw Sigmoid Probability", f"{result['probability']:.4f}")
-                                st.metric("Raw Model Output", f"{result['raw_score']:.4f}")
-                            with col2:
-                                st.info("""
-                                **How it works:**
-                                - Audio is resampled to 16kHz
-                                - Processed by SpecTTTra transformer
-                                - Output < 0.5 = Real, ≥ 0.5 = Fake
-                                """)
-                        
-                        # Progress bars for visualization
-                        st.markdown("### 📈 Confidence Visualization")
-                        if result["classification"] == "Real":
-                            st.progress(result["confidence"], text=f"Human Confidence: {result['confidence']:.1%}")
-                        else:
-                            st.progress(result["confidence"], text=f"AI Confidence: {result['confidence']:.1%}")
-                    
-                    else:
-                        # Multiple files analysis
-                        progress_bar = st.progress(0, text="🧠 Analyzing files...")
-                        results = []
-                        start_time = time.time()
-                        
-                        for i, file in enumerate(uploaded_files):
-                            progress = (i + 1) / len(uploaded_files)
-                            progress_bar.progress(progress, text=f"🧠 Analyzing file {i+1}/{len(uploaded_files)}: {file.name}")
-                            
-                            try:
-                                file_result = classify_audio_file(file, st.session_state.model)
-                                file_result["filename"] = file.name
-                                file_result["success"] = True
-                                results.append(file_result)
-                            except Exception as e:
-                                results.append({
-                                    "filename": file.name,
-                                    "success": False,
-                                    "error": str(e)
-                                })
-                        
-                        total_time = time.time() - start_time
-                        progress_bar.progress(1.0, text="✅ Analysis complete!")
-                        
-                        # Display bulk results
-                        st.markdown("### 🎯 Bulk Analysis Results")
-                        
-                        # Summary metrics
-                        successful = sum(1 for r in results if r["success"])
-                        failed = len(results) - successful
-                        
-                        col1, col2, col3, col4 = st.columns(4)
-                        with col1:
-                            st.metric("📁 Total Files", len(results))
-                        with col2:
-                            st.metric("✅ Successful", successful)
-                        with col3:
-                            st.metric("❌ Failed", failed)
-                        with col4:
-                            st.metric("⏱️ Total Time", f"{total_time:.1f}s")
-                        
-                        # Results breakdown
-                        if successful > 0:
-                            real_count = sum(1 for r in results if r.get("success") and r.get("classification") == "Real")
-                            fake_count = sum(1 for r in results if r.get("success") and r.get("classification") == "Fake")
-                            
-                            col1, col2 = st.columns(2)
-                            with col1:
-                                st.markdown(f"""
-                                <div class="result-box real-music">
-                                    <h4>🎤 Human Music: {real_count} files</h4>
-                                </div>
-                                """, unsafe_allow_html=True)
-                            with col2:
-                                st.markdown(f"""
-                                <div class="result-box fake-music">
-                                    <h4>🤖 AI Music: {fake_count} files</h4>
-                                </div>
-                                """, unsafe_allow_html=True)
-                        
-                        # Detailed results table
-                        st.markdown("### 📋 Detailed Results")
-                        
-                        for i, result in enumerate(results, 1):
-                            with st.expander(f"📄 {i}. {result['filename']}" + 
-                                           (" ✅" if result["success"] else " ❌")):
-                                if result["success"]:
-                                    col1, col2, col3 = st.columns(3)
-                                    with col1:
-                                        st.metric("Classification", result["classification"])
-                                    with col2:
-                                        st.metric("Confidence", f"{result['confidence']:.1%}")
-                                    with col3:
-                                        st.metric("Duration", f"{result['duration']:.1f}s")
-                                    
-                                    # Confidence bar
-                                    if result["classification"] == "Real":
-                                        st.progress(result["confidence"], 
-                                                  text=f"Human Confidence: {result['confidence']:.1%}")
-                                    else:
-                                        st.progress(result["confidence"], 
-                                                  text=f"AI Confidence: {result['confidence']:.1%}")
-                                else:
-                                    st.error(f"❌ Analysis failed: {result['error']}")
-                    
-                except Exception as e:
-                    st.error(f"❌ Error analyzing audio: {str(e)}")
-    
-    else:
-        # Model not loaded
-        st.markdown("""
-        <div class="info-box">
-            <h3>🚀 Getting Started</h3>
-            <p>Click "Load AI Model" in the sidebar to begin analyzing audio files.</p>
-            <p><strong>Note:</strong> The first load may take a moment as the model downloads.</p>
-        </div>
-        """, unsafe_allow_html=True)
-        
-        # Show supported formats
-        st.markdown("### 📁 Supported Audio Formats")
-        col1, col2, col3, col4, col5 = st.columns(5)
-        formats = [
-            ("🎵 WAV", ".wav"),
-            ("🎧 MP3", ".mp3"),
-            ("💿 FLAC", ".flac"),
-            ("📱 M4A", ".m4a"),
-            ("🎼 OGG", ".ogg")
-        ]
-        
-        for i, (icon_name, ext) in enumerate(formats):
-            with [col1, col2, col3, col4, col5][i]:
-                st.markdown(f"""
-                <div class="metric-card">
-                    <h4>{icon_name}</h4>
-                    <p>{ext}</p>
-                </div>
-                """, unsafe_allow_html=True)
-
-    # Footer
-    st.markdown("---")
-    st.markdown("""
-    <div style="text-align: center; color: #666; padding: 1rem;">
-        <p>🎵 <strong>Madverse Music</strong> - AI Music Detection Technology</p>
-        <p>Visit <a href="https://madverse.co" target="_blank">madverse.co</a> for more AI music tools</p>
-        <p><em>This tool is designed for research and testing purposes.</em></p>
-    </div>
-    """, unsafe_allow_html=True)
-
-if __name__ == "__main__":
-    main()