TTS web interface connected to the new RVC pipeline

8b8c37b8 · salma · e2d1bde7 · 8b8c37b8 · 8b8c37b8 · 8b8c37b8
Commit 8b8c37b8 authored Dec 02, 2025 by salma
Showing with 250 additions and 0 deletions

Dockerfile TTS_web_interface/Dockerfile +13 -0

captain-definition TTS_web_interface/captain-definition +4 -0

main.py TTS_web_interface/main.py +229 -0

requirements.txt TTS_web_interface/requirements.txt +4 -0

No files found.
--- a/TTS_web_interface/Dockerfile
+++ b/TTS_web_interface/Dockerfile
+FROM python:3.11-slim
+WORKDIR /app
+# Install dependencies
+COPY requirements.txt .
+RUN pip install --no-cache-dir -r requirements.txt
+# Copy application
+COPY main.py .
+# Run the application
+CMD ["python", "main.py"]
\ No newline at end of file
--- a/TTS_web_interface/captain-definition
+++ b/TTS_web_interface/captain-definition
+{
+  "schemaVersion": 2,
+  "dockerfilePath": "./Dockerfile"
+}
--- a/TTS_web_interface/main.py
+++ b/TTS_web_interface/main.py
+from fastapi import FastAPI, HTTPException
+from fastapi.responses import HTMLResponse, StreamingResponse
+from pydantic import BaseModel
+import httpx
+import io
+import os
+app = FastAPI()
+# ==========================================
+# 1. HTML FRONTEND
+# ==========================================
+HTML_CONTENT = """
+<!DOCTYPE html>
+<html lang="ar" dir="rtl">
+<head>
+    <meta charset="UTF-8">
+    <meta name="viewport" content="width=device-width, initial-scale=1.0">
+    <title>تحويل النص إلى صوت</title>
+    <style>
+        * { margin: 0; padding: 0; box-sizing: border-box; }
+        body {
+            font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
+            background: linear-gradient(135deg, #667eea 0%, #764ba2 100%);
+            min-height: 100vh;
+            display: flex; justify-content: center; align-items: center; padding: 20px;
+        }
+        .container {
+            background: white; border-radius: 20px;
+            box-shadow: 0 20px 60px rgba(0, 0, 0, 0.3);
+            padding: 40px; max-width: 600px; width: 100%;
+        }
+        h1 { color: #667eea; margin-bottom: 30px; text-align: center; font-size: 2em; }
+        textarea {
+            width: 100%; min-height: 200px; padding: 15px;
+            border: 2px solid #e0e0e0; border-radius: 10px;
+            font-size: 16px; font-family: inherit; resize: vertical; margin-bottom: 20px;
+        }
+        textarea:focus { outline: none; border-color: #667eea; }
+        /* NEW: Speed Control Styles */
+        .control-group { margin-bottom: 20px; display: flex; align-items: center; gap: 10px; }
+        .control-group label { font-weight: bold; color: #444; }
+        .control-group input { 
+            padding: 10px; border: 2px solid #e0e0e0; border-radius: 10px; width: 100px; font-size: 16px; 
+        }
+        .button-group { display: flex; gap: 15px; margin-top: 10px; }
+        button {
+            flex: 1; padding: 15px 30px; border: none; border-radius: 10px;
+            font-size: 16px; font-weight: bold; cursor: pointer; transition: all 0.3s;
+        }
+        #generateBtn { background: linear-gradient(135deg, #667eea 0%, #764ba2 100%); color: white; }
+        #generateBtn:disabled { background: #ccc; cursor: not-allowed; }
+        #downloadBtn { background: #10b981; color: white; display: none; }
+        audio { width: 100%; margin-top: 20px; display: none; }
+        .status { margin-top: 20px; padding: 15px; border-radius: 10px; text-align: center; font-weight: bold; display: none; }
+        .status.loading { background: #fef3c7; color: #92400e; display: block; }
+        .status.success { background: #d1fae5; color: #065f46; display: block; }
+        .status.error { background: #fee2e2; color: #991b1b; display: block; }
+        .spinner {
+            display: inline-block; width: 16px; height: 16px;
+            border: 3px solid #92400e; border-top-color: transparent;
+            border-radius: 50%; animation: spin 1s linear infinite; margin-left: 10px;
+        }
+        @keyframes spin { to { transform: rotate(360deg); } }
+    </style>
+</head>
+<body>
+    <div class="container">
+        <h1>🎙️ تحويل النص إلى صوت</h1>
+        <textarea id="textInput" placeholder="اكتب النص العربي هنا...">مرحباً بك في خدمة تحويل النص إلى صوت.</textarea>
+        <!-- NEW: Speed Input -->
+        <div class="control-group">
+            <label for="speedInput">سرعة الصوت (1.0 = عادي):</label>
+            <input type="number" id="speedInput" value="1.0" min="0.25" max="4.0" step="0.05">
+        </div>
+        <div class="button-group">
+            <button id="generateBtn" onclick="generateAudio()">توليد الصوت</button>
+            <button id="downloadBtn" onclick="downloadAudio()">تحميل الملف</button>
+        </div>
+        <audio id="audioPlayer" controls></audio>
+        <div id="status" class="status"></div>
+    </div>
+    <script>
+        let audioBlob = null;
+        async function generateAudio() {
+            const text = document.getElementById('textInput').value.trim();
+            // NEW: Get speed value
+            const speed = parseFloat(document.getElementById('speedInput').value); 
+            const generateBtn = document.getElementById('generateBtn');
+            const downloadBtn = document.getElementById('downloadBtn');
+            const audioPlayer = document.getElementById('audioPlayer');
+            if (!text) {
+                showStatus('error', 'الرجاء إدخال نص');
+                return;
+            }
+            generateBtn.disabled = true;
+            downloadBtn.style.display = 'none';
+            audioPlayer.style.display = 'none';
+            audioPlayer.src = '';
+            showStatus('loading', 'جاري الاتصال بالسيرفر...');
+            try {
+                // NEW: Send text AND speed
+                const response = await fetch('/synthesize', {
+                    method: 'POST',
+                    headers: { 'Content-Type': 'application/json' },
+                    body: JSON.stringify({ 
+                        text: text,
+                        speed: speed 
+                    })
+                });
+                if (!response.ok) {
+                    const err = await response.json();
+                    throw new Error(err.detail || 'فشل في توليد الصوت');
+                }
+                audioBlob = await response.blob();
+                const audioUrl = URL.createObjectURL(audioBlob);
+                audioPlayer.src = audioUrl;
+                audioPlayer.style.display = 'block';
+                audioPlayer.play();
+                showStatus('success', '✓ تم التوليد بنجاح!');
+                downloadBtn.style.display = 'block';
+            } catch (error) {
+                showStatus('error', '❌ حدث خطأ: ' + error.message);
+            } finally {
+                generateBtn.disabled = false;
+            }
+        }
+        function downloadAudio() {
+            if (!audioBlob) return;
+            const url = URL.createObjectURL(audioBlob);
+            const a = document.createElement('a');
+            a.href = url;
+            a.download = 'arabic_speech.wav';
+            document.body.appendChild(a);
+            a.click();
+            document.body.removeChild(a);
+        }
+        function showStatus(type, message) {
+            const status = document.getElementById('status');
+            status.className = 'status ' + type;
+            if (type === 'loading') status.innerHTML = message + '<span class="spinner"></span>';
+            else status.textContent = message;
+        }
+    </script>
+</body>
+</html>
+"""
+# ==========================================
+# 2. FASTAPI APP
+# ==========================================
+class TextInput(BaseModel):
+    text: str
+    speed: float = 1.0  # <--- NEW: Added speed parameter
+@app.get("/", response_class=HTMLResponse)
+async def read_root():
+    return HTML_CONTENT
+@app.get("/health")
+async def health_check():
+    return {"status": "healthy"}
+@app.post("/synthesize")
+async def synthesize(input_data: TextInput):
+    """
+    Directly forwards text to the external API without modification.
+    """
+    try:
+        # Configuration
+        external_url = "http://ec2-18-193-226-85.eu-central-1.compute.amazonaws.com:5000/generate_audio"
+        # Raw payload
+        payload = {
+            "text": input_data.text,
+            "speed": input_data.speed # <--- NEW: Forwarding speed to External API
+        }
+        # Call the external API (Timeout set to 60s for stability)
+        async with httpx.AsyncClient(timeout=60.0) as client:
+            response = await client.post(external_url, json=payload)
+            if response.status_code != 200:
+                print(f"External API Error: {response.text}")
+                raise HTTPException(status_code=response.status_code, detail="External TTS API failed")
+            # Stream result
+            return StreamingResponse(
+                io.BytesIO(response.content),
+                media_type="audio/wav",
+                headers={
+                    "Content-Disposition": "attachment; filename=arabic_speech.wav"
+                }
+            )
+    except httpx.TimeoutException:
+        raise HTTPException(status_code=504, detail="TTS API took too long (Timeout)")
+    except Exception as e:
+        print(f"Server Error: {str(e)}")
+        raise HTTPException(status_code=500, detail=str(e))
+if __name__ == "__main__":
+    import uvicorn
+    print("Starting server on http://localhost:80")
+    uvicorn.run(app, host="0.0.0.0", port=80)
\ No newline at end of file
--- a/TTS_web_interface/requirements.txt
+++ b/TTS_web_interface/requirements.txt
+fastapi
+uvicorn[standard]
+httpx
+pydantic
\ No newline at end of file