Auto-deploy from GitHub: 898f681ae67425055f39b96fde956658f2fe0d29

.gitattributes

@@ -1,35 +0,0 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -0,0 +1,34 @@
+FROM python:3.10-slim
+
+# Set working directory
+WORKDIR /app
+
+# Install system dependencies
+RUN apt-get update && apt-get install -y \
+    ffmpeg \
+    git \
+    curl \
+    && rm -rf /var/lib/apt/lists/*
+
+# Copy requirements first for better caching
+COPY requirements.txt .
+
+# Install Python dependencies
+RUN pip install --no-cache-dir -r requirements.txt
+
+# Copy application files
+COPY . .
+
+# Create necessary directories
+RUN mkdir -p uploads temp_dir
+
+# Expose port
+EXPOSE 7860
+
+# Create a startup script
+RUN echo '#!/bin/bash\n\
+python worker.py &\n\
+python app.py' > /app/start.sh && chmod +x /app/start.sh
+
+# Run the application
+CMD ["/app/start.sh"]

README.md

@@ -1,10 +1,34 @@
 ---
-title: STT
-emoji: 📚
-colorFrom: gray
-colorTo: pink
+title: STT Audio Caption Generator
+emoji: 🎵
+colorFrom: blue
+colorTo: purple
 sdk: docker
 pinned: false
+license: mit
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# Audio Caption Generator
+
+A Python-based audio transcription service with a neobrutalist web interface.
+
+## Features
+- 🎵 Audio file upload via REST API
+- 🤖 Automatic STT processing using faster-whisper
+- 💾 SQLite database for queue management
+- 🎨 Neobrutalist UI with smooth animations
+- 🔄 Real-time status updates
+
+## Usage
+Access the web interface at the Space URL above.
+
+## API Endpoints
+- POST `/api/upload` - Upload audio file
+- GET `/api/files` - Get all files
+- GET `/api/files/<id>` - Get specific file
+
+## Supported Formats
+WAV, MP3, FLAC, OGG, M4A, AAC
+
+---
+*Auto-deployed from GitHub*

app.py

@@ -0,0 +1,124 @@
+from flask import Flask, request, jsonify, send_from_directory
+from flask_cors import CORS
+import sqlite3
+import os
+import uuid
+from datetime import datetime
+from werkzeug.utils import secure_filename
+
+app = Flask(__name__)
+CORS(app)
+
+UPLOAD_FOLDER = 'uploads'
+ALLOWED_EXTENSIONS = {'wav', 'mp3', 'flac', 'ogg', 'm4a', 'aac'}
+
+os.makedirs(UPLOAD_FOLDER, exist_ok=True)
+os.makedirs('temp_dir', exist_ok=True)
+
+def init_db():
+    conn = sqlite3.connect('audio_captions.db')
+    c = conn.cursor()
+    c.execute('''CREATE TABLE IF NOT EXISTS audio_files
+                 (id TEXT PRIMARY KEY,
+                  filename TEXT NOT NULL,
+                  filepath TEXT NOT NULL,
+                  status TEXT NOT NULL,
+                  caption TEXT,
+                  created_at TEXT NOT NULL,
+                  processed_at TEXT)''')
+    conn.commit()
+    conn.close()
+
+def allowed_file(filename):
+    return '.' in filename and filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS
+
+@app.route('/')
+def index():
+    return send_from_directory('.', 'index.html')
+
+@app.route('/api/upload', methods=['POST'])
+def upload_audio():
+    if 'audio' not in request.files:
+        return jsonify({'error': 'No audio file provided'}), 400
+    
+    file = request.files['audio']
+    
+    if file.filename == '':
+        return jsonify({'error': 'No file selected'}), 400
+    
+    if not allowed_file(file.filename):
+        return jsonify({'error': 'Invalid file type'}), 400
+    
+    file_id = str(uuid.uuid4())
+    filename = secure_filename(file.filename)
+    filepath = os.path.join(UPLOAD_FOLDER, f"{file_id}_{filename}")
+    file.save(filepath)
+    
+    conn = sqlite3.connect('audio_captions.db')
+    c = conn.cursor()
+    c.execute('''INSERT INTO audio_files 
+                 (id, filename, filepath, status, created_at)
+                 VALUES (?, ?, ?, ?, ?)''',
+              (file_id, filename, filepath, 'not_started', datetime.now().isoformat()))
+    conn.commit()
+    conn.close()
+    
+    return jsonify({
+        'id': file_id,
+        'filename': filename,
+        'status': 'not_started',
+        'message': 'File uploaded successfully'
+    }), 201
+
+@app.route('/api/files', methods=['GET'])
+def get_files():
+    conn = sqlite3.connect('audio_captions.db')
+    conn.row_factory = sqlite3.Row
+    c = conn.cursor()
+    c.execute('SELECT * FROM audio_files ORDER BY created_at DESC')
+    rows = c.fetchall()
+    conn.close()
+    
+    files = []
+    for row in rows:
+        files.append({
+            'id': row['id'],
+            'filename': row['filename'],
+            'status': row['status'],
+            'caption': row['caption'],
+            'created_at': row['created_at'],
+            'processed_at': row['processed_at']
+        })
+    
+    return jsonify(files)
+
+@app.route('/api/files/<file_id>', methods=['GET'])
+def get_file(file_id):
+    conn = sqlite3.connect('audio_captions.db')
+    conn.row_factory = sqlite3.Row
+    c = conn.cursor()
+    c.execute('SELECT * FROM audio_files WHERE id = ?', (file_id,))
+    row = c.fetchone()
+    conn.close()
+    
+    if row is None:
+        return jsonify({'error': 'File not found'}), 404
+    
+    return jsonify({
+        'id': row['id'],
+        'filename': row['filename'],
+        'status': row['status'],
+        'caption': row['caption'],
+        'created_at': row['created_at'],
+        'processed_at': row['processed_at']
+    })
+
+@app.route('/health', methods=['GET'])
+def health():
+    return jsonify({'status': 'healthy', 'service': 'audio-caption-generator'})
+
+if __name__ == '__main__':
+    init_db()
+    # Use PORT environment variable for Hugging Face compatibility
+    port = int(os.environ.get('PORT', 7860))
+    app.run(debug=False, host='0.0.0.0', port=port)

index.html

@@ -0,0 +1,546 @@
+<!DOCTYPE html>
+<html lang="en">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>Audio Caption Generator</title>
+    <style>
+        * {
+            margin: 0;
+            padding: 0;
+            box-sizing: border-box;
+        }
+
+        :root {
+            --bg: #0a0e27;
+            --surface: #141b3d;
+            --primary: #00ff88;
+            --secondary: #ff00ff;
+            --accent: #00d4ff;
+            --error: #ff1744;
+            --text: #ffffff;
+            --border: 4px;
+        }
+
+        body {
+            font-family: 'Space Grotesk', 'Courier New', monospace;
+            background: var(--bg);
+            color: var(--text);
+            min-height: 100vh;
+            overflow-x: hidden;
+            position: relative;
+        }
+
+        body::before {
+            content: '';
+            position: fixed;
+            top: 0;
+            left: 0;
+            width: 100%;
+            height: 100%;
+            background: 
+                radial-gradient(circle at 20% 50%, rgba(0, 255, 136, 0.1) 0%, transparent 50%),
+                radial-gradient(circle at 80% 80%, rgba(255, 0, 255, 0.1) 0%, transparent 50%),
+                radial-gradient(circle at 40% 20%, rgba(0, 212, 255, 0.1) 0%, transparent 50%);
+            pointer-events: none;
+            z-index: 0;
+        }
+
+        .container {
+            max-width: 1400px;
+            margin: 0 auto;
+            padding: 2rem;
+            position: relative;
+            z-index: 1;
+        }
+
+        header {
+            text-align: center;
+            margin-bottom: 3rem;
+            animation: slideDown 0.6s cubic-bezier(0.68, -0.55, 0.265, 1.55);
+        }
+
+        @keyframes slideDown {
+            from {
+                opacity: 0;
+                transform: translateY(-50px);
+            }
+            to {
+                opacity: 1;
+                transform: translateY(0);
+            }
+        }
+
+        h1 {
+            font-size: clamp(2rem, 5vw, 4rem);
+            font-weight: 900;
+            background: linear-gradient(135deg, var(--primary) 0%, var(--accent) 50%, var(--secondary) 100%);
+            -webkit-background-clip: text;
+            -webkit-text-fill-color: transparent;
+            background-clip: text;
+            text-transform: uppercase;
+            letter-spacing: -2px;
+            margin-bottom: 1rem;
+            position: relative;
+            display: inline-block;
+        }
+
+        h1::after {
+            content: '';
+            position: absolute;
+            bottom: -10px;
+            left: 50%;
+            transform: translateX(-50%);
+            width: 60%;
+            height: 6px;
+            background: linear-gradient(90deg, transparent, var(--primary), transparent);
+            animation: glow 2s ease-in-out infinite;
+        }
+
+        @keyframes glow {
+            0%, 100% { opacity: 0.5; }
+            50% { opacity: 1; }
+        }
+
+        .subtitle {
+            font-size: 1.2rem;
+            color: var(--accent);
+            letter-spacing: 2px;
+        }
+
+        .upload-section {
+            background: var(--surface);
+            border: var(--border) solid var(--primary);
+            box-shadow: 8px 8px 0 var(--primary);
+            padding: 2rem;
+            margin-bottom: 3rem;
+            position: relative;
+            transition: all 0.3s ease;
+            animation: slideUp 0.6s cubic-bezier(0.68, -0.55, 0.265, 1.55) 0.2s both;
+        }
+
+        @keyframes slideUp {
+            from {
+                opacity: 0;
+                transform: translateY(50px);
+            }
+            to {
+                opacity: 1;
+                transform: translateY(0);
+            }
+        }
+
+        .upload-section:hover {
+            transform: translate(-2px, -2px);
+            box-shadow: 12px 12px 0 var(--primary);
+        }
+
+        .upload-zone {
+            border: 3px dashed var(--accent);
+            padding: 3rem;
+            text-align: center;
+            cursor: pointer;
+            transition: all 0.3s ease;
+            background: rgba(0, 212, 255, 0.05);
+        }
+
+        .upload-zone:hover {
+            background: rgba(0, 212, 255, 0.1);
+            border-color: var(--primary);
+        }
+
+        .upload-zone.dragging {
+            background: rgba(0, 255, 136, 0.2);
+            border-color: var(--primary);
+            transform: scale(1.02);
+        }
+
+        input[type="file"] {
+            display: none;
+        }
+
+        .btn {
+            background: var(--primary);
+            color: var(--bg);
+            border: var(--border) solid var(--bg);
+            padding: 1rem 2rem;
+            font-size: 1.1rem;
+            font-weight: 900;
+            text-transform: uppercase;
+            cursor: pointer;
+            transition: all 0.2s ease;
+            box-shadow: 4px 4px 0 var(--bg);
+            letter-spacing: 1px;
+        }
+
+        .btn:hover {
+            transform: translate(-2px, -2px);
+            box-shadow: 6px 6px 0 var(--bg);
+        }
+
+        .btn:active {
+            transform: translate(2px, 2px);
+            box-shadow: 2px 2px 0 var(--bg);
+        }
+
+        .btn-secondary {
+            background: var(--accent);
+        }
+
+        .table-section {
+            animation: slideUp 0.6s cubic-bezier(0.68, -0.55, 0.265, 1.55) 0.4s both;
+        }
+
+        .table-wrapper {
+            overflow-x: auto;
+            background: var(--surface);
+            border: var(--border) solid var(--secondary);
+            box-shadow: 8px 8px 0 var(--secondary);
+        }
+
+        table {
+            width: 100%;
+            border-collapse: collapse;
+        }
+
+        thead {
+            background: linear-gradient(135deg, var(--primary), var(--accent));
+        }
+
+        th {
+            padding: 1.5rem 1rem;
+            text-align: left;
+            font-weight: 900;
+            text-transform: uppercase;
+            letter-spacing: 1px;
+            color: var(--bg);
+            border-right: 3px solid var(--bg);
+        }
+
+        th:last-child {
+            border-right: none;
+        }
+
+        tbody tr {
+            border-bottom: 2px solid rgba(0, 212, 255, 0.2);
+            transition: all 0.3s ease;
+            animation: fadeIn 0.5s ease;
+        }
+
+        @keyframes fadeIn {
+            from { opacity: 0; }
+            to { opacity: 1; }
+        }
+
+        tbody tr:hover {
+            background: rgba(0, 255, 136, 0.1);
+            transform: translateX(5px);
+        }
+
+        td {
+            padding: 1.5rem 1rem;
+            color: var(--text);
+        }
+
+        .status {
+            display: inline-block;
+            padding: 0.5rem 1rem;
+            border: 3px solid;
+            font-weight: 900;
+            text-transform: uppercase;
+            font-size: 0.85rem;
+            letter-spacing: 1px;
+        }
+
+        .status-not_started {
+            background: var(--bg);
+            border-color: var(--accent);
+            color: var(--accent);
+        }
+
+        .status-processing {
+            background: var(--bg);
+            border-color: var(--primary);
+            color: var(--primary);
+            animation: pulse 1.5s ease-in-out infinite;
+        }
+
+        @keyframes pulse {
+            0%, 100% { opacity: 1; }
+            50% { opacity: 0.6; }
+        }
+
+        .status-completed {
+            background: var(--primary);
+            border-color: var(--primary);
+            color: var(--bg);
+        }
+
+        .status-failed {
+            background: var(--error);
+            border-color: var(--error);
+            color: var(--text);
+        }
+
+        .caption-cell {
+            max-width: 400px;
+            overflow: hidden;
+            text-overflow: ellipsis;
+            white-space: nowrap;
+        }
+
+        .caption-cell:hover {
+            white-space: normal;
+            cursor: pointer;
+        }
+
+        .empty-state {
+            text-align: center;
+            padding: 4rem 2rem;
+            color: var(--accent);
+            font-size: 1.2rem;
+        }
+
+        .refresh-btn {
+            position: fixed;
+            bottom: 2rem;
+            right: 2rem;
+            width: 60px;
+            height: 60px;
+            border-radius: 50%;
+            background: var(--secondary);
+            border: var(--border) solid var(--bg);
+            box-shadow: 4px 4px 0 var(--bg);
+            cursor: pointer;
+            transition: all 0.3s ease;
+            display: flex;
+            align-items: center;
+            justify-content: center;
+            font-size: 1.5rem;
+            z-index: 1000;
+        }
+
+        .refresh-btn:hover {
+            transform: rotate(180deg) scale(1.1);
+            box-shadow: 6px 6px 0 var(--bg);
+        }
+
+        @media (max-width: 768px) {
+            .container {
+                padding: 1rem;
+            }
+
+            .upload-section, .table-wrapper {
+                box-shadow: 4px 4px 0 var(--primary);
+            }
+
+            th, td {
+                padding: 1rem 0.5rem;
+                font-size: 0.9rem;
+            }
+
+            .caption-cell {
+                max-width: 200px;
+            }
+        }
+
+        .notification {
+            position: fixed;
+            top: 2rem;
+            right: 2rem;
+            padding: 1.5rem 2rem;
+            background: var(--primary);
+            color: var(--bg);
+            border: var(--border) solid var(--bg);
+            box-shadow: 6px 6px 0 var(--bg);
+            font-weight: 900;
+            z-index: 2000;
+            animation: slideInRight 0.5s ease, slideOutRight 0.5s ease 3.5s;
+        }
+
+        @keyframes slideInRight {
+            from {
+                transform: translateX(400px);
+                opacity: 0;
+            }
+            to {
+                transform: translateX(0);
+                opacity: 1;
+            }
+        }
+
+        @keyframes slideOutRight {
+            to {
+                transform: translateX(400px);
+                opacity: 0;
+            }
+        }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <header>
+            <h1>Audio Caption Generator</h1>
+            <p class="subtitle">Transcribe • Process • Analyze</p>
+        </header>
+
+        <div class="upload-section">
+            <h2 style="margin-bottom: 1.5rem; color: var(--primary);">Upload Audio File</h2>
+            <div class="upload-zone" id="uploadZone">
+                <p style="font-size: 1.2rem; margin-bottom: 1rem;">📁 Drop audio file here or click to browse</p>
+                <p style="color: var(--accent); font-size: 0.9rem;">Supported: WAV, MP3, FLAC, OGG, M4A, AAC</p>
+                <input type="file" id="audioFile" accept=".wav,.mp3,.flac,.ogg,.m4a,.aac">
+            </div>
+            <button class="btn" id="uploadBtn" style="margin-top: 1.5rem; width: 100%;">
+                🚀 Upload & Process
+            </button>
+        </div>
+
+        <div class="table-section">
+            <h2 style="margin-bottom: 1.5rem; color: var(--secondary);">Processing Queue</h2>
+            <div class="table-wrapper">
+                <table>
+                    <thead>
+                        <tr>
+                            <th>Filename</th>
+                            <th>Status</th>
+                            <th>Caption</th>
+                            <th>Created</th>
+                            <th>Processed</th>
+                        </tr>
+                    </thead>
+                    <tbody id="filesTable">
+                        <tr>
+                            <td colspan="5" class="empty-state">No files uploaded yet. Start by uploading an audio file!</td>
+                        </tr>
+                    </tbody>
+                </table>
+            </div>
+        </div>
+    </div>
+
+    <button class="refresh-btn" id="refreshBtn" title="Refresh">🔄</button>
+
+    <script>
+        const API_URL = 'http://localhost:5000/api';
+        let selectedFile = null;
+
+        // Upload zone interactions
+        const uploadZone = document.getElementById('uploadZone');
+        const fileInput = document.getElementById('audioFile');
+
+        uploadZone.addEventListener('click', () => fileInput.click());
+
+        uploadZone.addEventListener('dragover', (e) => {
+            e.preventDefault();
+            uploadZone.classList.add('dragging');
+        });
+
+        uploadZone.addEventListener('dragleave', () => {
+            uploadZone.classList.remove('dragging');
+        });
+
+        uploadZone.addEventListener('drop', (e) => {
+            e.preventDefault();
+            uploadZone.classList.remove('dragging');
+            const file = e.dataTransfer.files[0];
+            if (file) {
+                fileInput.files = e.dataTransfer.files;
+                selectedFile = file;
+                showNotification(`Selected: ${file.name}`);
+            }
+        });
+
+        fileInput.addEventListener('change', (e) => {
+            selectedFile = e.target.files[0];
+            if (selectedFile) {
+                showNotification(`Selected: ${selectedFile.name}`);
+            }
+        });
+
+        // Upload button
+        document.getElementById('uploadBtn').addEventListener('click', async () => {
+            if (!selectedFile) {
+                showNotification('Please select a file first!', 'error');
+                return;
+            }
+
+            const formData = new FormData();
+            formData.append('audio', selectedFile);
+
+            try {
+                const response = await fetch(`${API_URL}/upload`, {
+                    method: 'POST',
+                    body: formData
+                });
+
+                const data = await response.json();
+                
+                if (response.ok) {
+                    showNotification('File uploaded successfully! 🎉');
+                    selectedFile = null;
+                    fileInput.value = '';
+                    loadFiles();
+                } else {
+                    showNotification(data.error || 'Upload failed', 'error');
+                }
+            } catch (error) {
+                showNotification('Network error: ' + error.message, 'error');
+            }
+        });
+
+        // Load files
+        async function loadFiles() {
+            try {
+                const response = await fetch(`${API_URL}/files`);
+                const files = await response.json();
+                
+                const tbody = document.getElementById('filesTable');
+                
+                if (files.length === 0) {
+                    tbody.innerHTML = '<tr><td colspan="5" class="empty-state">No files uploaded yet. Start by uploading an audio file!</td></tr>';
+                    return;
+                }
+                
+                tbody.innerHTML = files.map(file => `
+                    <tr>
+                        <td><strong>${file.filename}</strong></td>
+                        <td><span class="status status-${file.status}">${file.status.replace('_', ' ')}</span></td>
+                        <td class="caption-cell" title="${file.caption || 'N/A'}">${file.caption || '—'}</td>
+                        <td>${new Date(file.created_at).toLocaleString()}</td>
+                        <td>${file.processed_at ? new Date(file.processed_at).toLocaleString() : '—'}</td>
+                    </tr>
+                `).join('');
+            } catch (error) {
+                console.error('Error loading files:', error);
+            }
+        }
+
+        // Refresh button
+        document.getElementById('refreshBtn').addEventListener('click', loadFiles);
+
+        // Auto-refresh every 3 seconds
+        setInterval(loadFiles, 3000);
+
+        // Show notification
+        function showNotification(message, type = 'success') {
+            const notification = document.createElement('div');
+            notification.className = 'notification';
+            notification.textContent = message;
+            
+            if (type === 'error') {
+                notification.style.background = 'var(--error)';
+            }
+            
+            document.body.appendChild(notification);
+            
+            setTimeout(() => {
+                notification.remove();
+            }, 4000);
+        }
+
+        // Initial load
+        loadFiles();
+    </script>
+</body>
+</html>

requirements.txt

@@ -0,0 +1,6 @@
+Flask==3.0.0
+flask-cors==4.0.0
+werkzeug==3.0.1
+
+#STT
+git+https://github.com/jebin2/STT.git#egg=stt-runner[fasterwhisper]

worker.py

@@ -0,0 +1,125 @@
+import sqlite3
+import time
+import os
+import subprocess
+import json
+import shlex
+from datetime import datetime
+
+CWD = "./"
+PYTHON_PATH = "stt-transcribe"
+STT_MODEL_NAME = "fasterwhispher"
+POLL_INTERVAL = 5  # seconds
+
+def process_audio(file_id, filepath):
+    """Process audio file using STT and return the transcription"""
+    try:
+        print(f"Processing file: {filepath}")
+        
+        # Run STT command
+        command = f"""cd {CWD} && {PYTHON_PATH} --input {shlex.quote(os.path.abspath(filepath))} --model {STT_MODEL_NAME}"""
+        
+        subprocess.run(
+            command,
+            shell=True,
+            executable="/bin/bash",
+            check=True,
+            cwd=CWD,
+            env={
+                **os.environ,
+                'PYTHONUNBUFFERED': '1',
+                'CUDA_LAUNCH_BLOCKING': '1',
+                'USE_CPU_IF_POSSIBLE': 'true'
+            }
+        )
+        
+        # Read transcription result
+        output_path = f'{CWD}/temp_dir/output_transcription.json'
+        with open(output_path, 'r') as file:
+            result = json.loads(file.read().strip())
+        
+        # Extract caption text (adjust based on your actual output format)
+        caption = result.get('text', '') or result.get('transcription', '') or str(result)
+        
+        return caption, None
+        
+    except Exception as e:
+        print(f"Error processing file {file_id}: {str(e)}")
+        return None, str(e)
+
+def update_status(file_id, status, caption=None, error=None):
+    """Update the status of a file in the database"""
+    conn = sqlite3.connect('audio_captions.db')
+    c = conn.cursor()
+    
+    if status == 'completed':
+        c.execute('''UPDATE audio_files 
+                     SET status = ?, caption = ?, processed_at = ?
+                     WHERE id = ?''',
+                  (status, caption, datetime.now().isoformat(), file_id))
+    elif status == 'failed':
+        c.execute('''UPDATE audio_files 
+                     SET status = ?, caption = ?, processed_at = ?
+                     WHERE id = ?''',
+                  (status, f"Error: {error}", datetime.now().isoformat(), file_id))
+    else:
+        c.execute('UPDATE audio_files SET status = ? WHERE id = ?', (status, file_id))
+    
+    conn.commit()
+    conn.close()
+
+def worker_loop():
+    """Main worker loop that processes audio files"""
+    print("STT Worker started. Polling for new audio files...")
+    
+    while True:
+        try:
+            # Get next unprocessed file
+            conn = sqlite3.connect('audio_captions.db')
+            conn.row_factory = sqlite3.Row
+            c = conn.cursor()
+            c.execute('''SELECT * FROM audio_files 
+                         WHERE status = 'not_started' 
+                         ORDER BY created_at ASC 
+                         LIMIT 1''')
+            row = c.fetchone()
+            conn.close()
+            
+            if row:
+                file_id = row['id']
+                filepath = row['filepath']
+                filename = row['filename']
+                
+                print(f"\n{'='*60}")
+                print(f"Processing: {filename}")
+                print(f"ID: {file_id}")
+                print(f"{'='*60}")
+                
+                # Update status to processing
+                update_status(file_id, 'processing')
+                
+                # Process the audio file
+                caption, error = process_audio(file_id, filepath)
+                
+                if caption:
+                    print(f"✓ Successfully processed: {filename}")
+                    print(f"Caption: {caption[:100]}...")
+                    update_status(file_id, 'completed', caption=caption)
+                else:
+                    print(f"✗ Failed to process: {filename}")
+                    print(f"Error: {error}")
+                    update_status(file_id, 'failed', error=error)
+            else:
+                # No files to process, sleep for a bit
+                time.sleep(POLL_INTERVAL)
+                
+        except Exception as e:
+            print(f"Worker error: {str(e)}")
+            time.sleep(POLL_INTERVAL)
+
+if __name__ == '__main__':
+    # Initialize database if it doesn't exist
+    if not os.path.exists('audio_captions.db'):
+        print("Database not found. Please run app.py first to initialize.")
+    else:
+        worker_loop()