Upload 4 files

Dockerfile (1)

@@ -0,0 +1,41 @@
+# ── Base image ────────────────────────────────────────────────────────
+FROM python:3.10-slim
+
+# ── System deps (audio libs required by TTS / soundfile) ──────────────
+RUN apt-get update && apt-get install -y --no-install-recommends \
+        build-essential \
+        libsndfile1 \
+        ffmpeg \
+        git \
+    && rm -rf /var/lib/apt/lists/*
+
+# ── HuggingFace Spaces: non-root user uid=1000 ────────────────────────
+RUN useradd -m -u 1000 user
+USER user
+
+ENV HOME=/home/user \
+    PATH=/home/user/.local/bin:$PATH \
+    # Accept Coqui TOS automatically
+    COQUI_TOS_AGREED=1 \
+    # Keep HF model cache inside the container's writable home
+    HF_HOME=/home/user/.cache/huggingface \
+    # Avoid tokeniser parallelism warnings
+    TOKENIZERS_PARALLELISM=false
+
+WORKDIR $HOME/app
+
+# ── Python deps ───────────────────────────────────────────────────────
+COPY --chown=user requirements.txt .
+RUN pip install --no-cache-dir --upgrade pip \
+ && pip install --no-cache-dir -r requirements.txt
+
+# ── App code ─────────────────────────────────────────────────────────
+COPY --chown=user app.py .
+
+# ── Runtime dirs (writable by user 1000) ─────────────────────────────
+RUN mkdir -p /home/user/app/voice_library /home/user/app/outputs
+
+# ── HuggingFace Spaces listens on 7860 ───────────────────────────────
+EXPOSE 7860
+
+CMD ["python", "app.py"]

README (9).md

@@ -0,0 +1,31 @@
+---
+title: XTTS Voice Studio
+emoji: 🎙️
+colorFrom: yellow
+colorTo: orange
+sdk: docker
+app_port: 7860
+pinned: false
+---
+
+# XTTS v2 Voice Studio
+
+A multilingual text-to-speech studio powered by **Coqui XTTS v2**, served via **FastAPI** (no Gradio).
+
+## Features
+- 16 supported languages (Arabic, English, French, …)
+- Voice cloning from uploaded audio samples
+- Persistent voice library
+- Generation history with playback & download
+- Fully custom React UI
+
+## Usage
+1. Upload a reference audio clip (WAV / MP3 / FLAC, ≥ 6 s recommended).
+2. Type your text and pick a language.
+3. Adjust advanced parameters if needed.
+4. Click **⚡ توليد الصوت** and wait — CPU inference takes ~30–90 s per request.
+
+## Notes
+- Running on **CPU**; generation is slower than GPU but fully functional.
+- The XTTS v2 model (~1.8 GB) is downloaded on first startup and cached.
+- `COQUI_TOS_AGREED=1` is set automatically — by using this Space you agree to the [Coqui TTS terms](https://coqui.ai/cpml).

app (5).py

@@ -0,0 +1,312 @@
+# ╔══════════════════════════════════════════════════════════════════╗
+# ║          XTTS v2 Advanced Voice Studio — HuggingFace Space       ║
+# ╚══════════════════════════════════════════════════════════════════╝
+
+import os, sys, time, json, uuid, shutil, threading
+
+import uvicorn
+from fastapi import FastAPI, Form, File, UploadFile, HTTPException
+from fastapi.responses import HTMLResponse, FileResponse, JSONResponse
+from fastapi.middleware.cors import CORSMiddleware
+import torch
+
+# ── Env ──────────────────────────────────────────────────────────────
+os.environ["COQUI_TOS_AGREED"] = "1"
+# Point HF cache to a writable directory inside the Space
+os.environ.setdefault("HF_HOME", "/home/user/.cache/huggingface")
+
+VOICE_LIB    = "/home/user/app/voice_library"
+OUTPUT_DIR   = "/home/user/app/outputs"
+HISTORY_FILE = "/home/user/app/history.json"
+
+for d in [VOICE_LIB, OUTPUT_DIR]:
+    os.makedirs(d, exist_ok=True)
+
+# ── Load TTS on CPU ──────────────────────────────────────────────────
+from TTS.api import TTS
+
+# HF Spaces free tier is CPU-only; force CPU explicitly
+device = "cpu"
+
+print(f"[*] Loading XTTS v2 on {device.upper()} …")
+xtts_engine = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to(device)
+print("[✓] Model ready.")
+
+# ── History helpers ──────────────────────────────────────────────────
+def load_history():
+    if os.path.exists(HISTORY_FILE):
+        try:
+            return json.load(open(HISTORY_FILE))
+        except Exception:
+            pass
+    return []
+
+def save_history(h):
+    json.dump(h, open(HISTORY_FILE, "w"), ensure_ascii=False, indent=2)
+
+# ── FastAPI app ──────────────────────────────────────────────────────
+app = FastAPI(title="XTTS Studio")
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"], allow_methods=["*"], allow_headers=["*"]
+)
+
+LANGUAGES = {
+    "ar": "العربية",  "en": "English",   "es": "Español",  "fr": "Français",
+    "de": "Deutsch",  "it": "Italiano",  "pt": "Português","ru": "Русский",
+    "zh-cn": "中文", "ja": "日本語",    "ko": "한국어",   "tr": "Türkçe",
+    "nl": "Nederlands","pl": "Polski",   "cs": "Čeština",  "hi": "हिन्दी",
+}
+
+# ══════════════════════════════════════════════════════════════════════
+# HTML / React Frontend
+# ══════════════════════════════════════════════════════════════════════
+HTML = r"""<!DOCTYPE html>
+<html lang="ar" dir="rtl">
+<head>
+<meta charset="UTF-8">
+<meta name="viewport" content="width=device-width, initial-scale=1.0">
+<title>XTTS Voice Studio</title>
+<link href="https://fonts.googleapis.com/css2?family=IBM+Plex+Mono:wght@400;600&family=IBM+Plex+Sans+Arabic:wght@300;400;600;700&display=swap" rel="stylesheet">
+<script src="https://cdn.tailwindcss.com"></script>
+<script src="https://unpkg.com/react@18/umd/react.production.min.js"></script>
+<script src="https://unpkg.com/react-dom@18/umd/react-dom.production.min.js"></script>
+<script src="https://unpkg.com/@babel/standalone/babel.min.js"></script>
+<style>
+  :root { --bg: #0b0c0f; --surface: #13151a; --border: #1f2330; --amber: #f5a623; --amber-dim:#a06a10; --green: #3ddc84; --red: #ff5252; --text: #e8eaf0; --muted: #6b7280; --mono: 'IBM Plex Mono', monospace; --sans: 'IBM Plex Sans Arabic', sans-serif; }
+  * { box-sizing: border-box; } body { margin: 0; background: var(--bg); color: var(--text); font-family: var(--sans); min-height: 100vh; }
+  ::-webkit-scrollbar { width: 5px; } ::-webkit-scrollbar-track { background: var(--surface); } ::-webkit-scrollbar-thumb { background: var(--border); border-radius: 4px; }
+  body::before { content: ""; position: fixed; inset: 0; pointer-events: none; z-index: 0; background: repeating-linear-gradient(0deg, transparent, transparent 39px, rgba(255,255,255,.02) 40px); }
+  .card { background: var(--surface); border: 1px solid var(--border); border-radius: 12px; } .amber { color: var(--amber); } .tag { font-family: var(--mono); font-size: 10px; letter-spacing: .12em; text-transform: uppercase; color: var(--muted); }
+  input[type=range] { -webkit-appearance: none; width: 100%; height: 3px; background: var(--border); border-radius: 2px; outline: none; } input[type=range]::-webkit-slider-thumb { -webkit-appearance: none; width: 14px; height: 14px; background: var(--amber); border-radius: 50%; cursor: pointer; transition: transform .15s; } input[type=range]::-webkit-slider-thumb:hover { transform: scale(1.3); }
+  select, textarea { background: var(--bg); border: 1px solid var(--border); color: var(--text); border-radius: 6px; padding: 6px 10px; font-family: var(--sans); outline: none; transition: border-color .2s; } textarea { resize: vertical; padding: 12px; width: 100%; } select:focus, textarea:focus { border-color: var(--amber); }
+  .file-drop { border: 2px dashed var(--border); border-radius: 8px; padding: 16px; text-align: center; cursor: pointer; transition: border-color .2s, background .2s; position: relative; } .file-drop:hover, .file-drop.active { border-color: var(--amber); background: rgba(245,166,35,.05); } .file-drop input { position: absolute; inset: 0; opacity: 0; cursor: pointer; }
+  .btn-primary { background: var(--amber); color: #000; font-weight: 700; border: none; border-radius: 8px; padding: 14px 24px; cursor: pointer; font-family: var(--sans); font-size: 15px; width: 100%; transition: opacity .2s, transform .1s; } .btn-primary:hover { opacity: .9; } .btn-primary:active { transform: scale(.98); } .btn-primary:disabled { opacity: .4; cursor: not-allowed; }
+  .btn-ghost { background: transparent; border: 1px solid var(--border); color: var(--muted); border-radius: 6px; padding: 6px 12px; cursor: pointer; font-size: 12px; transition: color .2s, border-color .2s; } .btn-ghost:hover { color: var(--text); border-color: var(--muted); }
+  .badge { display: inline-flex; align-items: center; gap: 4px; background: rgba(245,166,35,.12); border: 1px solid rgba(245,166,35,.3); color: var(--amber); border-radius: 20px; padding: 2px 10px; font-size: 11px; font-family: var(--mono); } .badge.green { background: rgba(61,220,132,.1); border-color: rgba(61,220,132,.3); color: var(--green); } .badge.red { background: rgba(255,82,82,.1); border-color: rgba(255,82,82,.3); color: var(--red); }
+  .waveform { display: flex; align-items: center; gap: 3px; height: 28px; } .waveform span { flex: 1; background: var(--amber); border-radius: 2px; animation: wave 1s ease-in-out infinite; opacity: .7; } @keyframes wave { 0%,100%{height:4px} 50%{height:24px} } .waveform span:nth-child(2){animation-delay:.1s} .waveform span:nth-child(3){animation-delay:.2s} .waveform span:nth-child(4){animation-delay:.3s} .waveform span:nth-child(5){animation-delay:.2s} .waveform span:nth-child(6){animation-delay:.1s} .waveform span:nth-child(7){animation-delay:.05s}
+  .tab { cursor:pointer; padding:8px 16px; border-radius:6px; font-size:13px; color:var(--muted); transition:all .2s; } .tab.active { background:rgba(245,166,35,.15); color:var(--amber); } .tab:hover:not(.active) { color:var(--text); }
+  audio { width:100%; accent-color:var(--amber); } audio::-webkit-media-controls-panel { background:var(--surface); }
+  .history-row { display:flex; align-items:center; gap:12px; padding:10px 14px; border-radius:8px; border:1px solid var(--border); background:var(--bg); transition:border-color .2s; } .history-row:hover { border-color: var(--amber-dim); }
+  .param-row { display:grid; grid-template-columns:140px 1fr 48px; align-items:center; gap:12px; } .param-label { font-size:12px; color:var(--muted); font-family:var(--mono); } .param-val { font-size:13px; color:var(--amber); font-family:var(--mono); text-align:right; }
+  .voice-card { padding:10px 14px; border-radius:8px; border:1px solid var(--border); background:var(--bg); cursor:pointer; transition:all .2s; display:flex; align-items:center; justify-content:space-between; } .voice-card:hover { border-color:var(--amber-dim); } .voice-card.selected { border-color:var(--amber); background:rgba(245,166,35,.06); }
+</style>
+</head>
+<body>
+<div id="root"></div>
+<script type="text/babel">
+const { useState, useEffect, useRef, useCallback } = React;
+const device = "DEVICE_PLACEHOLDER";
+const fmt = v => parseFloat(v).toFixed(2);
+const apiPost = (url, body) => fetch(url, { method:"POST", body }).then(r => { if (!r.ok) return r.json().then(e => { throw new Error(e.detail || "Server error"); }); return r.json(); });
+
+function Slider({ label, min, max, step, value, onChange }) {
+  return ( <div className="param-row"><span className="param-label">{label}</span><input type="range" min={min} max={max} step={step} value={value} onChange={e => onChange(parseFloat(e.target.value))} /><span className="param-val">{fmt(value)}</span></div> );
+}
+
+function FileZone({ label, file, onFile }) {
+  const [active, setActive] = useState(false);
+  return ( <div><div className="tag mb-1">{label}</div><div className={`file-drop ${active?"active":""}`} onDragOver={e=>{e.preventDefault();setActive(true)}} onDragLeave={()=>setActive(false)} onDrop={e=>{e.preventDefault();setActive(false);onFile(e.dataTransfer.files[0]);}}> <input type="file" accept="audio/*" onChange={e=>onFile(e.target.files[0])} /> {file ? <span style={{color:"var(--green)",fontSize:12}}>✓ {file.name}</span> : <span style={{color:"var(--muted)",fontSize:12}}>اسحب ملفاً أو انقر</span>} </div></div> );
+}
+
+function WaveAnim() { return <div className="waveform">{[1,2,3,4,5,6,7].map(i=><span key={i}/>)}</div>; }
+
+function App() {
+  const [tab, setTab] = useState("generate");
+  const [text, setText] = useState(""); const [lang, setLang] = useState("ar"); const [file1, setFile1] = useState(null); const [file2, setFile2] = useState(null);
+  const [temperature, setTemp] = useState(0.75); const [speed, setSpeed] = useState(1.0); const [topK, setTopK] = useState(50); const [topP, setTopP] = useState(0.85); const [repPenalty, setRepPenalty] = useState(5.0); const [splitText, setSplitText] = useState(true);
+  const [status, setStatus] = useState("idle"); const [statusMsg, setStatusMsg] = useState(""); const [audioUrl, setAudioUrl] = useState(null); const [audioFilename, setAudioFilename] = useState(null);
+  const [history, setHistory] = useState([]); const [voices, setVoices] = useState([]); const [selVoice, setSelVoice] = useState(null); const [saveName, setSaveName] = useState(""); const [saveStatus, setSaveStatus] = useState("");
+  const languages = LANGUAGES_JSON;
+
+  useEffect(() => { fetch("/history").then(r=>r.json()).then(setHistory).catch(()=>{}); fetch("/voices").then(r=>r.json()).then(setVoices).catch(()=>{}); }, []);
+
+  const generate = async () => {
+    if (!text.trim()) return setStatusMsg("أدخل النص أولاً.");
+    if (!file1 && !selVoice) return setStatusMsg("يجب تحديد عينة صوتية أو اختيار صوت من المكتبة.");
+    setStatus("running"); setStatusMsg(""); setAudioUrl(null);
+    const fd = new FormData();
+    fd.append("text", text); fd.append("language", lang); fd.append("temperature", temperature); fd.append("speed", speed); fd.append("top_k", topK); fd.append("top_p", topP); fd.append("repetition_penalty", repPenalty); fd.append("enable_text_splitting", splitText);
+    if (file1) fd.append("files", file1); if (file2) fd.append("files", file2); if (selVoice) fd.append("voice_name", selVoice);
+    try { const data = await apiPost("/generate", fd); setAudioUrl(`/audio/${data.filename}`); setAudioFilename(data.filename); setStatus("done"); fetch("/history").then(r=>r.json()).then(setHistory).catch(()=>{}); } catch(e) { setStatus("error"); setStatusMsg(e.message); }
+  };
+
+  const saveVoice = async () => {
+    if (!saveName.trim() || !file1) return;
+    const fd = new FormData(); fd.append("name", saveName.trim()); fd.append("file", file1); if (file2) fd.append("file2", file2);
+    try { await apiPost("/voices/save", fd); setSaveStatus("✓ تم الحفظ"); fetch("/voices").then(r=>r.json()).then(setVoices).catch(()=>{}); setTimeout(()=>setSaveStatus(""),2000); } catch(e) { setSaveStatus("خطأ: " + e.message); }
+  };
+
+  const deleteVoice = async name => { await fetch(`/voices/${name}`, {method:"DELETE"}); setVoices(v => v.filter(x=>x!==name)); if (selVoice===name) setSelVoice(null); };
+  const isRTL = ["ar","fa","he","ur"].includes(lang);
+
+  return (
+    <div style={{maxWidth:720,margin:"0 auto",padding:"24px 16px",position:"relative",zIndex:1}}>
+      <div style={{marginBottom:28,textAlign:"center"}}>
+        <div className="tag" style={{marginBottom:6}}>XTTS V2 MULTILINGUAL</div>
+        <h1 style={{margin:0,fontSize:26,fontWeight:700,letterSpacing:"-.02em"}}><span className="amber">Voice</span> Studio</h1>
+        <div style={{marginTop:8,display:"flex",justifyContent:"center",gap:6}}><span className={`badge ${device==="cuda"?"green":"red"}`}>● {device.toUpperCase()}</span><span className="badge">HuggingFace Space</span></div>
+      </div>
+      <div style={{display:"flex",gap:4,marginBottom:20,background:"var(--surface)",border:"1px solid var(--border)",borderRadius:8,padding:4}}>
+        {["generate","library","history"].map(t=>( <div key={t} className={`tab${tab===t?" active":""}`} onClick={()=>setTab(t)} style={{flex:1,textAlign:"center"}}>{t==="generate"?"⚡ توليد":t==="library"?"📚 مكتبة الأصوات":"🕘 السجل"}</div> ))}
+      </div>
+      {tab==="generate" && (
+        <div style={{display:"flex",flexDirection:"column",gap:14}}>
+          <div className="card" style={{padding:16}}><div style={{display:"flex",justifyContent:"space-between",alignItems:"center",marginBottom:10}}><div className="tag">النص المراد تحويله</div><select value={lang} onChange={e=>setLang(e.target.value)}>{Object.entries(languages).map(([k,v])=><option key={k} value={k}>{v}</option>)}</select></div><textarea dir={isRTL?"rtl":"ltr"} rows={5} placeholder={isRTL?"أدخل النص هنا…":"Enter text here…"} value={text} onChange={e=>setText(e.target.value)} /></div>
+          <div className="card" style={{padding:16}}><div className="tag" style={{marginBottom:10}}>المرجع الصوتي</div><div style={{display:"grid",gridTemplateColumns:"1fr 1fr",gap:12}}><FileZone label="عينة 1 (مطلوبة)" file={file1} onFile={setFile1} /><FileZone label="عينة 2 (اختيارية)" file={file2} onFile={setFile2} /></div>
+            {voices.length>0 && ( <div style={{marginTop:12}}><div className="tag" style={{marginBottom:8}}>أو من المكتبة</div><div style={{display:"flex",flexWrap:"wrap",gap:8}}>{voices.map(v=>( <div key={v} style={{padding:"6px 14px",borderRadius:20,cursor:"pointer",fontSize:12,border:"1px solid",borderColor:selVoice===v?"var(--amber)":"var(--border)",color:selVoice===v?"var(--amber)":"var(--muted)",background:selVoice===v?"rgba(245,166,35,.08)":"transparent",transition:"all .2s"}} onClick={()=>setSelVoice(selVoice===v?null:v)}>{v}</div> ))}</div></div> )}
+            {file1 && ( <div style={{marginTop:12,display:"flex",gap:8,alignItems:"center"}}><input type="text" placeholder="اسم الصوت للحفظ…" value={saveName} onChange={e=>setSaveName(e.target.value)} style={{flex:1,background:"var(--bg)",border:"1px solid var(--border)",color:"var(--text)",borderRadius:6,padding:"6px 10px",fontSize:12,fontFamily:"var(--sans)",outline:"none"}} /><button className="btn-ghost" onClick={saveVoice} style={{whiteSpace:"nowrap"}}>حفظ في المكتبة</button>{saveStatus && <span style={{fontSize:12,color:"var(--green)"}}>{saveStatus}</span>}</div> )}
+          </div>
+          <div className="card" style={{padding:16}}><details><summary style={{cursor:"pointer",listStyle:"none",display:"flex",justifyContent:"space-between",alignItems:"center",userSelect:"none"}}><div className="tag">المعاملات المتقدمة</div><span style={{fontSize:11,color:"var(--muted)"}}>انقر للتوسيع ▾</span></summary><div style={{marginTop:14,display:"flex",flexDirection:"column",gap:14}}><Slider label="temperature" min={0.05} max={1.0} step={0.05} value={temperature} onChange={setTemp} /><Slider label="speed" min={0.5} max={2.0} step={0.05} value={speed} onChange={setSpeed} /><Slider label="top_k" min={1} max={100} step={1} value={topK} onChange={setTopK} /><Slider label="top_p" min={0.1} max={1.0} step={0.05} value={topP} onChange={setTopP} /><Slider label="rep_penalty" min={1.0} max={10.0} step={0.5} value={repPenalty} onChange={setRepPenalty} /></div></details></div>
+          {statusMsg && <div style={{padding:"10px 14px",borderRadius:8,fontSize:13,background:"rgba(255,82,82,.08)",border:"1px solid rgba(255,82,82,.3)",color:"var(--red)"}}>{statusMsg}</div>}
+          <button className="btn-primary" onClick={generate} disabled={status==="running"}>{status==="running" ? <span style={{display:"flex",alignItems:"center",justifyContent:"center",gap:10}}><WaveAnim/> جاري التوليد…</span> : "⚡ توليد الصوت"}</button>
+          {audioUrl && ( <div className="card" style={{padding:16}}><div style={{display:"flex",justifyContent:"space-between",alignItems:"center",marginBottom:12}}><div style={{display:"flex",alignItems:"center",gap:8}}><span className="badge green">✓ تم التوليد</span></div><a href={audioUrl} download={audioFilename} style={{textDecoration:"none"}}><button className="btn-ghost">⬇ تحميل</button></a></div><audio src={audioUrl} controls autoPlay /></div> )}
+        </div>
+      )}
+      {tab==="library" && (
+        <div style={{display:"flex",flexDirection:"column",gap:12}}>
+          <div className="card" style={{padding:16}}><div className="tag" style={{marginBottom:8}}>إضافة صوت جديد للمكتبة</div><div style={{display:"grid",gridTemplateColumns:"1fr 1fr",gap:12,marginBottom:12}}><FileZone label="عينة 1" file={file1} onFile={setFile1} /><FileZone label="عينة 2 (اختياري)" file={file2} onFile={setFile2} /></div><div style={{display:"flex",gap:8}}><input type="text" placeholder="اسم الصوت…" value={saveName} onChange={e=>setSaveName(e.target.value)} style={{flex:1,background:"var(--bg)",border:"1px solid var(--border)",color:"var(--text)",borderRadius:6,padding:"8px 12px",fontSize:13,fontFamily:"var(--sans)",outline:"none"}} /><button className="btn-primary" style={{width:"auto",padding:"8px 20px"}} onClick={saveVoice}>حفظ</button></div>{saveStatus && <div style={{marginTop:8,fontSize:12,color:"var(--green)"}}>{saveStatus}</div>}</div>
+          {voices.length===0 ? <div style={{textAlign:"center",color:"var(--muted)",padding:"40px 0",fontSize:14}}>لا توجد أصوات محفوظة بعد.</div> : voices.map(v=>( <div key={v} className="voice-card" onClick={()=>{setSelVoice(selVoice===v?null:v);setTab("generate");}}><div><div style={{fontSize:14,fontWeight:600}}>{v}</div><div style={{fontSize:11,color:"var(--muted)",marginTop:2,fontFamily:"var(--mono)"}}>{selVoice===v?"● محدد":"انقر للاستخدام"}</div></div><button className="btn-ghost" onClick={e=>{e.stopPropagation();deleteVoice(v);}}>حذف</button></div> ))}
+        </div>
+      )}
+      {tab==="history" && (
+        <div style={{display:"flex",flexDirection:"column",gap:10}}>
+          {history.length===0 ? <div style={{textAlign:"center",color:"var(--muted)",padding:"40px 0",fontSize:14}}>السجل فارغ.</div> : [...history].reverse().map((h,i)=>( <div key={i} className="history-row"><div style={{flex:1,minWidth:0}}><div style={{fontSize:12,color:"var(--text)",overflow:"hidden",textOverflow:"ellipsis",whiteSpace:"nowrap",direction:"rtl"}}>{h.text}</div><div style={{display:"flex",gap:8,marginTop:4,flexWrap:"wrap"}}><span className="badge">{h.language}</span><span style={{fontSize:10,color:"var(--muted)",fontFamily:"var(--mono)"}}>{new Date(h.ts*1000).toLocaleString("ar-EG")}</span></div></div><div style={{display:"flex",gap:6,flexShrink:0}}><button className="btn-ghost" onClick={()=>{setAudioUrl(`/audio/${h.filename}`);setAudioFilename(h.filename);setTab("generate");setStatus("done");}}></button><a href={`/audio/${h.filename}`} download={h.filename} style={{textDecoration:"none"}}><button className="btn-ghost">⬇</button></a></div></div> ))}
+        </div>
+      )}
+    </div>
+  );
+}
+ReactDOM.createRoot(document.getElementById("root")).render(<App/>);
+</script>
+</body>
+</html>
+"""
+
+# ══════════════════════════════════════════════════════════════════════
+# Routes
+# ══════════════════════════════════════════════════════════════════════
+
+@app.get("/", response_class=HTMLResponse)
+async def ui():
+    page = (
+        HTML
+        .replace("LANGUAGES_JSON", json.dumps(LANGUAGES, ensure_ascii=False))
+        .replace("DEVICE_PLACEHOLDER", device)
+    )
+    return page
+
+
+@app.post("/generate")
+async def generate(
+    text: str = Form(...),
+    language: str = Form("ar"),
+    temperature: float = Form(0.75),
+    speed: float = Form(1.0),
+    top_k: int = Form(50),
+    top_p: float = Form(0.85),
+    repetition_penalty: float = Form(5.0),
+    enable_text_splitting: bool = Form(True),
+    voice_name: str = Form(None),
+    files: list[UploadFile] = File(default=[]),
+):
+    if not text.strip():
+        raise HTTPException(400, "النص فارغ.")
+
+    ref_paths, tmp_files = [], []
+    for f in files:
+        path = f"/tmp/ref_{uuid.uuid4().hex}_{f.filename}"
+        with open(path, "wb") as buf:
+            shutil.copyfileobj(f.file, buf)
+        ref_paths.append(path)
+        tmp_files.append(path)
+
+    if voice_name:
+        lib_dir = os.path.join(VOICE_LIB, voice_name)
+        if os.path.isdir(lib_dir):
+            ref_paths += [
+                os.path.join(lib_dir, fn)
+                for fn in os.listdir(lib_dir)
+                if fn.lower().endswith((".wav", ".mp3", ".flac", ".ogg"))
+            ]
+
+    if not ref_paths:
+        raise HTTPException(400, "يجب تحديد عينة صوتية مرجعية.")
+
+    out_name = f"gen_{uuid.uuid4().hex[:8]}.wav"
+    out_path = os.path.join(OUTPUT_DIR, out_name)
+
+    try:
+        xtts_engine.tts_to_file(
+            text=text,
+            speaker_wav=ref_paths,
+            language=language,
+            file_path=out_path,
+            temperature=float(temperature),
+            speed=float(speed),
+            top_k=int(top_k),
+            top_p=float(top_p),
+            repetition_penalty=float(repetition_penalty),
+            enable_text_splitting=bool(enable_text_splitting),
+        )
+    finally:
+        for p in tmp_files:
+            try:
+                os.remove(p)
+            except Exception:
+                pass
+
+    hist = load_history()
+    hist.append({"filename": out_name, "text": text[:120], "language": language, "ts": int(time.time())})
+    save_history(hist)
+    return {"filename": out_name}
+
+
+@app.get("/audio/{filename}")
+def get_audio(filename: str):
+    path = os.path.join(OUTPUT_DIR, filename)
+    if not os.path.exists(path):
+        raise HTTPException(404, "File not found.")
+    return FileResponse(path, media_type="audio/wav")
+
+
+@app.get("/history")
+def get_history():
+    return JSONResponse(load_history())
+
+
+@app.get("/voices")
+def list_voices():
+    if not os.path.isdir(VOICE_LIB):
+        return []
+    return [d for d in os.listdir(VOICE_LIB) if os.path.isdir(os.path.join(VOICE_LIB, d))]
+
+
+@app.post("/voices/save")
+async def save_voice(
+    name: str = Form(...),
+    file: UploadFile = File(...),
+    file2: UploadFile = File(default=None),
+):
+    safe = name.strip().replace("/", "_").replace("..", "_")
+    lib_dir = os.path.join(VOICE_LIB, safe)
+    os.makedirs(lib_dir, exist_ok=True)
+    for f in ([file, file2] if file2 else [file]):
+        with open(os.path.join(lib_dir, f.filename), "wb") as buf:
+            shutil.copyfileobj(f.file, buf)
+    return {"name": safe}
+
+
+@app.delete("/voices/{name}")
+def delete_voice(name: str):
+    lib_dir = os.path.join(VOICE_LIB, name)
+    if os.path.isdir(lib_dir):
+        shutil.rmtree(lib_dir)
+    return {"deleted": name}
+
+
+# ════════════════��═════════════════════════════════════════════════════
+# Entry point (used by Dockerfile CMD)
+# ══════════════════════════════════════════════════════════════════════
+if __name__ == "__main__":
+    uvicorn.run(app, host="0.0.0.0", port=7860, log_level="info")

requirements (6).txt

@@ -0,0 +1,11 @@
+# TTS engine
+coqui-tts>=0.22.0
+
+# Web framework
+fastapi>=0.111.0
+uvicorn[standard]>=0.29.0
+python-multipart>=0.0.9
+
+# PyTorch CPU-only wheel (much smaller image, sufficient for inference)
+torch>=2.2.0
+torchaudio>=2.2.0