品質の良い音声クローンWebアプリをローカル環境(Win/Mac)に構築するガイド 2026.2版

"""
Qwen3-TTS Voice Cloning - ローカルGUIアプリ
Streamlit版 | マルチプラットフォーム対応

対応環境:
  - Mac Apple Silicon (M1/M2/M3/M4) : mlx_audio バックエンド
  - Windows CUDA                     : qwen_tts + PyTorch CUDA バックエンド
  - Windows CPU                      : qwen_tts + PyTorch CPU バックエンド

起動方法:
    source .venv/bin/activate   # Mac
    .venv\\Scripts\\activate     # Windows
    streamlit run app.py
"""

import streamlit as st
import os, sys, shutil, tempfile, time, wave, re, subprocess, platform
from pathlib import Path
from datetime import datetime

# ========== ページ設定 ==========
st.set_page_config(
    page_title="Qwen3-TTS Voice Cloning",
    page_icon="🎙️",
    layout="wide",
    initial_sidebar_state="expanded",
)

st.markdown("""
<style>
.main-header{background:linear-gradient(135deg,#1a1a2e,#16213e,#0f3460);
 padding:2rem;border-radius:12px;margin-bottom:1.5rem;text-align:center}
.main-header h1{color:#e94560;font-size:2.2rem;margin:0;font-weight:700}
.main-header p{color:#a8b2d8;margin:.5rem 0 0;font-size:1rem}
.stButton>button{background:linear-gradient(135deg,#e94560,#c23152);color:white;
 border:none;border-radius:8px;padding:.7rem 2rem;font-size:1rem;font-weight:600;width:100%}
.stButton>button:hover{background:linear-gradient(135deg,#c23152,#a01f3f);
 transform:translateY(-1px);box-shadow:0 4px 12px rgba(233,69,96,.3)}
.status-bar{background:#1a1a2e;color:#a8b2d8;padding:.8rem 1rem;border-radius:8px;
 font-family:monospace;font-size:.85rem;min-height:3rem}
.hint-box{background:#fff3cd;border:1px solid #ffc107;border-radius:8px;
 padding:.8rem;font-size:.85rem;color:#856404}
.env-badge{display:inline-block;padding:2px 10px;border-radius:12px;
 font-size:.8rem;font-weight:600;color:white}
</style>
""", unsafe_allow_html=True)

st.markdown("""
<div class="main-header">
    <h1>🎙️ Qwen3-TTS Voice Cloning</h1>
    <p>ローカル完全オフライン - 3〜5秒のサンプル音声から声を複製</p>
</div>
""", unsafe_allow_html=True)

# ========== バックエンド検出 ==========
@st.cache_resource
def detect_backend():
    """利用可能なバックエンドを自動検出"""
    # 1. Mac MLX (Apple Silicon)
    try:
        from mlx_audio.tts.utils import load_model      # noqa
        from mlx_audio.tts.generate import generate_audio  # noqa
        return "mlx"
    except ImportError:
        pass
    # 2. PyTorch (Windows CUDA / CPU)
    try:
        import torch
        from qwen_tts import Qwen3TTSModel  # noqa
        if torch.cuda.is_available():
            return "cuda"
        return "cpu"
    except ImportError:
        pass
    return None

BACKEND = detect_backend()

# ========== パス設定 ==========
IS_MAC  = platform.system() == "Darwin"
IS_WIN  = platform.system() == "Windows"

if IS_MAC:
    INSTALL_DIR = Path.home() / "qwen3-tts-apple-silicon"
    MODELS_DIR  = INSTALL_DIR / "models"
    VOICES_DIR  = INSTALL_DIR / "voices"
    OUTPUT_DIR  = INSTALL_DIR / "outputs"
    SAMPLE_RATE = 24000
    # MLX版はフォルダ名に -8bit がつく
    MODEL_SUFFIX = "-8bit"
else:
    # Windows: app.py と同じ場所に models/ を置く想定
    INSTALL_DIR = Path(__file__).parent
    MODELS_DIR  = INSTALL_DIR / "models"
    VOICES_DIR  = INSTALL_DIR / "voices"
    OUTPUT_DIR  = INSTALL_DIR / "outputs"
    SAMPLE_RATE = 24000
    MODEL_SUFFIX = ""   # 公式モデルはそのままのフォルダ名

MODEL_FOLDERS = {
    ("1.7B", "Voice Cloning"):  f"Qwen3-TTS-12Hz-1.7B-Base{MODEL_SUFFIX}",
    ("1.7B", "Custom Voice"):   f"Qwen3-TTS-12Hz-1.7B-CustomVoice{MODEL_SUFFIX}",
    ("1.7B", "Voice Design"):   f"Qwen3-TTS-12Hz-1.7B-VoiceDesign{MODEL_SUFFIX}",
    ("0.6B", "Voice Cloning"):  f"Qwen3-TTS-12Hz-0.6B-Base{MODEL_SUFFIX}",
    ("0.6B", "Custom Voice"):   f"Qwen3-TTS-12Hz-0.6B-CustomVoice{MODEL_SUFFIX}",
    ("0.6B", "Voice Design"):   f"Qwen3-TTS-12Hz-0.6B-VoiceDesign{MODEL_SUFFIX}",
}

# ========== ユーティリティ ==========
def get_model_path(folder_name: str):
    """HuggingFace snapshots 構造にも対応"""
    p = MODELS_DIR / folder_name
    if not p.exists():
        return None
    snap = p / "snapshots"
    if snap.exists():
        subs = [f for f in snap.iterdir() if not f.name.startswith(".")]
        if subs:
            return str(subs[0])
    return str(p)

def convert_to_wav(src: str) -> str | None:
    if not os.path.exists(src):
        return None
    _, ext = os.path.splitext(src)
    if ext.lower() == ".wav":
        try:
            with wave.open(src, "rb") as f:
                if f.getnchannels() > 0:
                    return src
        except wave.Error:
            pass
    out = str(INSTALL_DIR / f"tmp_conv_{int(time.time())}.wav")
    cmd = ["ffmpeg", "-y", "-v", "error", "-i", src,
           "-ar", str(SAMPLE_RATE), "-ac", "1", "-c:a", "pcm_s16le", out]
    try:
        subprocess.run(cmd, check=True, stdout=subprocess.DEVNULL, stderr=subprocess.PIPE)
        return out
    except (subprocess.CalledProcessError, FileNotFoundError):
        return None

def get_saved_voices():
    if not VOICES_DIR.exists():
        return []
    return sorted([f.stem for f in VOICES_DIR.glob("*.wav")])

# ========== サイドバー ==========
with st.sidebar:
    st.markdown("### ⚙️ 設定")

    model_size  = st.selectbox("🤖 モデルサイズ", ["1.7B（高品質）", "0.6B（軽量）"])
    size_key    = "1.7B" if "1.7B" in model_size else "0.6B"
    mode_label  = st.selectbox("🎛️ モード", [
        "Voice Cloning（音声クローン）",
        "Custom Voice（プリセット話者）",
        "Voice Design（声質デザイン）",
    ])
    mode_short  = mode_label.split("（")[0]

    st.markdown("---")
    st.markdown("### 📊 システム情報")
    st.caption(f"OS: {platform.system()} {platform.machine()}")

    backend_labels = {
        "mlx":  ("🍎 MLX (Apple Silicon)", "#27ae60"),
        "cuda": ("⚡ PyTorch CUDA",         "#2980b9"),
        "cpu":  ("🖥️ PyTorch CPU",          "#e67e22"),
        None:   ("❌ バックエンドなし",       "#e74c3c"),
    }
    label, color = backend_labels[BACKEND]
    st.markdown(f'<span class="env-badge" style="background:{color}">{label}</span>',
                unsafe_allow_html=True)
    if BACKEND is None:
        st.error("バックエンドが見つかりません")
        if IS_MAC:
            st.code("pip install mlx-audio", language="bash")
        else:
            st.code("pip install qwen-tts\npip install torch", language="bash")

    current_folder = MODEL_FOLDERS.get((size_key, mode_short), "")
    current_model  = get_model_path(current_folder)
    if current_model:
        st.success(f"✓ モデル検出\n`{current_folder}`")
    else:
        st.warning(f"⚠ モデル未検出\n`{current_folder}`")
        with st.expander("ダウンロード方法"):
            hf_name = current_folder.replace("-8bit", "").replace(
                "Qwen3-TTS-12Hz-", "Qwen/Qwen3-TTS-12Hz-")
            st.code(
                f"huggingface-cli download {hf_name} \\\n"
                f"  --local-dir ./models/{current_folder}",
                language="bash",
            )

    st.markdown("---")
    if IS_WIN and BACKEND == "cpu":
        st.warning("⚠ CPU モード: 処理に数分かかります")
    elif IS_WIN and BACKEND == "cuda":
        try:
            import torch
            st.caption(f"CUDA: {torch.version.cuda}")
            st.caption(f"GPU: {torch.cuda.get_device_name(0)}")
        except Exception:
            pass

# ========== メインレイアウト ==========
col_left, col_right = st.columns([1, 1], gap="large")

# ============================
# 左カラム：入力
# ============================
with col_left:

    if "Voice Cloning" in mode_label:
        st.markdown("#### 🎤 参照音声（Reference Audio）")
        st.caption("クローンしたい声のサンプルをアップロード（3〜5秒推奨）")

        clone_src = st.radio("音声の取得方法",
                             ["新しくアップロード", "保存済みの声を使う"], horizontal=True)

        ref_audio_path = None
        ref_text_val   = ""

        if clone_src == "新しくアップロード":
            uploaded = st.file_uploader("参照音声", type=["wav","mp3","m4a","flac"],
                                        label_visibility="collapsed")
            if uploaded:
                st.audio(uploaded)
                suffix = Path(uploaded.name).suffix
                tmp = tempfile.NamedTemporaryFile(delete=False, suffix=suffix,
                                                  dir=str(INSTALL_DIR))
                tmp.write(uploaded.read())
                tmp.close()
                ref_audio_path = tmp.name
                st.success(f"✓ {uploaded.name}")
            else:
                st.markdown("""
                <div class="hint-box">💡 WAV推奨・3〜5秒・雑音なし・16kHz以上</div>
                """, unsafe_allow_html=True)
        else:
            saved = get_saved_voices()
            if saved:
                sel = st.selectbox("保存済みの声", saved)
                ref_audio_path = str(VOICES_DIR / f"{sel}.wav")
                txt = VOICES_DIR / f"{sel}.txt"
                if txt.exists():
                    ref_text_val = txt.read_text(encoding="utf-8").strip()
                st.audio(ref_audio_path)
            else:
                st.info("保存済みの声がありません。")

        st.markdown("")
        st.markdown("#### 📝 参照テキスト（Reference Text）")
        ref_text_val = st.text_area("参照テキスト", value=ref_text_val,
            placeholder="例: こんにちは。これはクローニングのテストです。",
            height=80, label_visibility="collapsed")
        use_xvector = st.checkbox("🔬 x-vector のみ（参照テキスト不要・品質低下あり）")
        if use_xvector:
            ref_text_val = "."

        save_voice = st.checkbox("✅ この声を名前をつけて保存する")
        voice_name = st.text_input("保存名", placeholder="例: Boss, Mom") if save_voice else ""

    elif "Custom Voice" in mode_label:
        st.markdown("#### 👤 話者の選択")
        spk_map = {
            "Japanese": ["Ono_Anna"],
            "English":  ["Ryan","Aiden","Ethan","Chelsie","Serena","Vivian"],
            "Chinese":  ["Vivian","Serena","Uncle_Fu","Dylan","Eric"],
            "Korean":   ["Sohee"],
        }
        lang     = st.selectbox("言語", list(spk_map.keys()))
        speaker  = st.selectbox("話者", spk_map[lang])
        st.markdown("#### 🎭 感情・口調")
        emotions = ["Normal tone","Sad and crying, speaking slowly",
                    "Excited and happy, speaking very fast",
                    "Angry and shouting","Whispering quietly"]
        preset   = st.selectbox("プリセット", emotions)
        custom_e = st.text_input("カスタム指定（英語）", placeholder="例: Calm and professional")
        instruct_val = custom_e.strip() or preset
        st.markdown("#### ⚡ 速度")
        speed_opt = st.select_slider("速度", [0.8,1.0,1.3], value=1.0,
            format_func=lambda x:{0.8:"遅(0.8x)",1.0:"普通(1.0x)",1.3:"速(1.3x)"}[x])
    else:
        st.markdown("#### 🎨 声質のデザイン")
        instruct_val = st.text_area("声質の説明（英語）",
            placeholder="例: A calm middle-aged Japanese male voice, speaking slowly.",
            height=100, label_visibility="collapsed")

    st.markdown("---")
    st.markdown("#### 💬 生成テキスト（Target Text）")
    target_text = st.text_area("生成テキスト",
        placeholder="例: 今日も一日、お疲れ様でした。",
        height=110, label_visibility="collapsed")

    # トーン調整（Voice Cloning モードのみ）
    tone_instruct = ""
    if "Voice Cloning" in mode_label:
        st.markdown("#### 🎭 トーン調整")
        tone_presets = {
            "通常（指定なし）":         "",
            "ゆっくり丁寧に":           "Speak slowly and politely.",
            "明るく元気に":             "Speak in a bright and energetic tone.",
            "落ち着いてナレーション風": "Speak calmly like a narrator.",
            "ニュースキャスター風":     "Speak clearly and professionally like a news anchor.",
            "悲しそうに・しんみりと":   "Speak sadly and solemnly.",
        }
        tone_label  = st.selectbox("話し方のスタイル", list(tone_presets.keys()),
                                   label_visibility="collapsed")
        tone_custom = st.text_input("カスタム指定（英語）",
                                    placeholder="例: Speak with excitement and enthusiasm.")
        tone_instruct = tone_custom.strip() or tone_presets[tone_label]

    generate_btn = st.button("🎵  Clone & Generate", use_container_width=True)

# ============================
# 右カラム：出力
# ============================
with col_right:
    st.markdown("#### 🔊 生成音声（Generated Audio）")
    audio_ph  = st.empty()
    status_ph = st.empty()
    log_ph    = st.empty()

    with audio_ph.container():
        st.markdown("""
        <div style="background:#f0f2f6;border:2px dashed #ccc;border-radius:10px;
                    padding:3rem;text-align:center;color:#999;">
            <div style="font-size:3rem;">🎧</div>
            <div>生成後にここに音声が表示されます</div>
        </div>""", unsafe_allow_html=True)

    with status_ph.container():
        st.markdown('<div class="status-bar">> 待機中...</div>',
                    unsafe_allow_html=True)

    st.markdown("---")
    with st.expander("📖 使い方", expanded=True):
        st.markdown("""
**Voice Cloning:** 参照音声をアップロード → 参照テキスト入力 → 生成テキスト入力 → Generate

**Custom Voice:** 話者・感情を選んで生成テキスト入力

**Voice Design:** 声の特徴を英語で記述して生成
        """)

# ========== 生成処理 ==========
def upd(msg):
    status_ph.markdown(f'<div class="status-bar">> {msg}</div>',
                       unsafe_allow_html=True)

if generate_btn:
    errs = []
    if not target_text.strip():
        errs.append("生成テキストを入力してください。")
    if "Voice Cloning" in mode_label and not ref_audio_path:
        errs.append("参照音声をアップロードしてください。")
    if BACKEND is None:
        errs.append("バックエンドがインストールされていません。")
    if not current_model:
        errs.append(f"モデルが見つかりません: `{current_folder}`")
    for e in errs:
        st.error(f"❌ {e}")

    if not errs:
        OUTPUT_DIR.mkdir(parents=True, exist_ok=True)
        ts        = datetime.now().strftime("%H-%M-%S")
        clean_t   = re.sub(r'[^\w\s-]', '', target_text)[:20].strip().replace(' ','_') or "audio"
        out_name  = f"{ts}_{clean_t}.wav"
        temp_dir  = str(INSTALL_DIR / f"tmp_{int(time.time())}")
        os.makedirs(temp_dir, exist_ok=True)

        try:
            # ===== MLX バックエンド（Mac Apple Silicon）=====
            if BACKEND == "mlx":
                from mlx_audio.tts.utils import load_model
                from mlx_audio.tts.generate import generate_audio

                cache_key = f"mlx_model_{current_folder}"
                if cache_key not in st.session_state:
                    upd(f"モデルを読み込んでいます（MLX）: {current_folder} ...")
                    st.session_state[cache_key] = load_model(current_model)
                model = st.session_state[cache_key]
                upd("音声を生成しています（MLX）...")

                if "Voice Cloning" in mode_label:
                    wav_path = convert_to_wav(ref_audio_path)
                    if not wav_path:
                        st.error("❌ 参照音声の変換失敗。ffmpegを確認してください。")
                        st.stop()
                    gen_kwargs = dict(
                        model=model, text=target_text,
                        ref_audio=wav_path, ref_text=ref_text_val,
                        output_path=temp_dir,
                    )
                    if tone_instruct:
                        gen_kwargs["instruct"] = tone_instruct
                    generate_audio(**gen_kwargs)
                    if save_voice and voice_name.strip():
                        VOICES_DIR.mkdir(exist_ok=True)
                        safe = re.sub(r'[^\w-]','',voice_name).strip()
                        shutil.copy(wav_path, str(VOICES_DIR/f"{safe}.wav"))
                        if ref_text_val:
                            (VOICES_DIR/f"{safe}.txt").write_text(ref_text_val, encoding="utf-8")
                    if wav_path != ref_audio_path and os.path.exists(wav_path):
                        os.remove(wav_path)
                elif "Custom Voice" in mode_label:
                    generate_audio(model=model, text=target_text,
                                   voice=speaker, instruct=instruct_val,
                                   speed=speed_opt, output_path=temp_dir)
                else:
                    generate_audio(model=model, text=target_text,
                                   instruct=instruct_val, output_path=temp_dir)

                src_wav = os.path.join(temp_dir, "audio_000.wav")

            # ===== PyTorch バックエンド（Windows CUDA / CPU）=====
            else:
                import torch
                import soundfile as sf
                from qwen_tts import Qwen3TTSModel

                device = "cuda" if BACKEND == "cuda" else "cpu"
                dtype  = torch.float16 if BACKEND == "cuda" else torch.float32

                cache_key = f"pt_model_{current_folder}"
                if cache_key not in st.session_state:
                    upd(f"モデルを読み込んでいます（{device.upper()}）: {current_folder} ...")
                    st.session_state[cache_key] = Qwen3TTSModel.from_pretrained(
                        current_model,
                        device_map=device,
                        dtype=dtype,
                        attn_implementation="sdpa",
                    )
                model = st.session_state[cache_key]
                upd(f"音声を生成しています（{device.upper()}）...")

                src_wav = os.path.join(temp_dir, "output.wav")

                if "Voice Cloning" in mode_label:
                    clone_kwargs = dict(
                        text=target_text,
                        ref_audio=ref_audio_path,
                        ref_text=ref_text_val if not use_xvector else None,
                    )
                    if tone_instruct:
                        clone_kwargs["instruct"] = tone_instruct
                    wavs, sr = model.generate_voice_clone(**clone_kwargs)
                    sf.write(src_wav, wavs[0], sr)
                    if save_voice and voice_name.strip():
                        VOICES_DIR.mkdir(exist_ok=True)
                        safe = re.sub(r'[^\w-]','',voice_name).strip()
                        shutil.copy(ref_audio_path, str(VOICES_DIR/f"{safe}.wav"))
                        if ref_text_val:
                            (VOICES_DIR/f"{safe}.txt").write_text(ref_text_val, encoding="utf-8")
                else:
                    # Custom Voice / Voice Design
                    wavs, sr = model.generate(text=target_text)
                    sf.write(src_wav, wavs[0], sr)

            # 出力先に移動
            final_dir = OUTPUT_DIR / mode_short.replace(" ","")
            final_dir.mkdir(parents=True, exist_ok=True)
            final_path = str(final_dir / out_name)
            if os.path.exists(src_wav):
                shutil.move(src_wav, final_path)
            shutil.rmtree(temp_dir, ignore_errors=True)

            upd(f"✓ 完了 → {final_path}")

            with audio_ph.container():
                st.success("✅ 生成完了！")
                with open(final_path, "rb") as f:
                    ab = f.read()
                st.audio(ab, format="audio/wav")
                st.download_button("⬇️ WAVをダウンロード", ab,
                                   file_name=out_name, mime="audio/wav",
                                   use_container_width=True)

            with log_ph.container():
                with st.expander("📋 生成パラメータ"):
                    p = {"モード": mode_label, "バックエンド": BACKEND,
                         "モデル": current_folder, "生成テキスト": target_text}
                    if "Voice Cloning" in mode_label:
                        p["参照テキスト"] = ref_text_val
                    elif "Custom Voice" in mode_label:
                        p["話者"] = speaker; p["感情"] = instruct_val; p["速度"] = speed_opt
                    else:
                        p["声質指定"] = instruct_val
                    st.json(p)

        except Exception as e:
            upd(f"❌ エラー: {e}")
            st.error(f"```\n{e}\n```")

st.markdown("---")
st.markdown("""
<div style="text-align:center;color:#888;font-size:.8rem">
    Qwen3-TTS ローカルGUI | Powered by Qwen Team (Alibaba Cloud) |
    <a href="https://github.com/QwenLM/Qwen3-TTS">GitHub</a>
</div>""", unsafe_allow_html=True)

"""
Qwen3-TTS Voice Cloning - ローカルGUIアプリ
Streamlit版 | マルチプラットフォーム対応

対応環境:
  - Mac Apple Silicon (M1/M2/M3/M4) : mlx_audio バックエンド
  - Windows CUDA                     : qwen_tts + PyTorch CUDA バックエンド
  - Windows CPU                      : qwen_tts + PyTorch CPU バックエンド

起動方法:
    source .venv/bin/activate   # Mac
    .venv\\Scripts\\activate     # Windows
    streamlit run app.py
"""

import streamlit as st
import os, sys, shutil, tempfile, time, wave, re, subprocess, platform
from pathlib import Path
from datetime import datetime

# ========== ページ設定 ==========
st.set_page_config(
    page_title="Qwen3-TTS Voice Cloning",
    page_icon="🎙️",
    layout="wide",
    initial_sidebar_state="expanded",
)

st.markdown("""
<style>
.main-header{background:linear-gradient(135deg,#1a1a2e,#16213e,#0f3460);
 padding:2rem;border-radius:12px;margin-bottom:1.5rem;text-align:center}
.main-header h1{color:#e94560;font-size:2.2rem;margin:0;font-weight:700}
.main-header p{color:#a8b2d8;margin:.5rem 0 0;font-size:1rem}
.stButton>button{background:linear-gradient(135deg,#e94560,#c23152);color:white;
 border:none;border-radius:8px;padding:.7rem 2rem;font-size:1rem;font-weight:600;width:100%}
.stButton>button:hover{background:linear-gradient(135deg,#c23152,#a01f3f);
 transform:translateY(-1px);box-shadow:0 4px 12px rgba(233,69,96,.3)}
.status-bar{background:#1a1a2e;color:#a8b2d8;padding:.8rem 1rem;border-radius:8px;
 font-family:monospace;font-size:.85rem;min-height:3rem}
.hint-box{background:#fff3cd;border:1px solid #ffc107;border-radius:8px;
 padding:.8rem;font-size:.85rem;color:#856404}
.env-badge{display:inline-block;padding:2px 10px;border-radius:12px;
 font-size:.8rem;font-weight:600;color:white}
</style>
""", unsafe_allow_html=True)

st.markdown("""
<div class="main-header">
    <h1>🎙️ Qwen3-TTS Voice Cloning</h1>
    <p>ローカル完全オフライン - 3〜5秒のサンプル音声から声を複製</p>
</div>
""", unsafe_allow_html=True)

# ========== バックエンド検出 ==========
@st.cache_resource
def detect_backend():
    """利用可能なバックエンドを自動検出"""
    # 1. Mac MLX (Apple Silicon)
    try:
        from mlx_audio.tts.utils import load_model      # noqa
        from mlx_audio.tts.generate import generate_audio  # noqa
        return "mlx"
    except ImportError:
        pass
    # 2. PyTorch (Windows CUDA / CPU)
    try:
        import torch
        from qwen_tts import Qwen3TTSModel  # noqa
        if torch.cuda.is_available():
            return "cuda"
        return "cpu"
    except ImportError:
        pass
    return None

BACKEND = detect_backend()

# ========== パス設定 ==========
IS_MAC  = platform.system() == "Darwin"
IS_WIN  = platform.system() == "Windows"

if IS_MAC:
    INSTALL_DIR = Path.home() / "qwen3-tts-apple-silicon"
    MODELS_DIR  = INSTALL_DIR / "models"
    VOICES_DIR  = INSTALL_DIR / "voices"
    OUTPUT_DIR  = INSTALL_DIR / "outputs"
    SAMPLE_RATE = 24000
    # MLX版はフォルダ名に -8bit がつく
    MODEL_SUFFIX = "-8bit"
else:
    # Windows: app.py と同じ場所に models/ を置く想定
    INSTALL_DIR = Path(__file__).parent
    MODELS_DIR  = INSTALL_DIR / "models"
    VOICES_DIR  = INSTALL_DIR / "voices"
    OUTPUT_DIR  = INSTALL_DIR / "outputs"
    SAMPLE_RATE = 24000
    MODEL_SUFFIX = ""   # 公式モデルはそのままのフォルダ名

MODEL_FOLDERS = {
    ("1.7B", "Voice Cloning"):  f"Qwen3-TTS-12Hz-1.7B-Base{MODEL_SUFFIX}",
    ("1.7B", "Custom Voice"):   f"Qwen3-TTS-12Hz-1.7B-CustomVoice{MODEL_SUFFIX}",
    ("1.7B", "Voice Design"):   f"Qwen3-TTS-12Hz-1.7B-VoiceDesign{MODEL_SUFFIX}",
    ("0.6B", "Voice Cloning"):  f"Qwen3-TTS-12Hz-0.6B-Base{MODEL_SUFFIX}",
    ("0.6B", "Custom Voice"):   f"Qwen3-TTS-12Hz-0.6B-CustomVoice{MODEL_SUFFIX}",
    ("0.6B", "Voice Design"):   f"Qwen3-TTS-12Hz-0.6B-VoiceDesign{MODEL_SUFFIX}",
}

# ========== ユーティリティ ==========
def get_model_path(folder_name: str):
    """HuggingFace snapshots 構造にも対応"""
    p = MODELS_DIR / folder_name
    if not p.exists():
        return None
    snap = p / "snapshots"
    if snap.exists():
        subs = [f for f in snap.iterdir() if not f.name.startswith(".")]
        if subs:
            return str(subs[0])
    return str(p)

def convert_to_wav(src: str) -> str | None:
    if not os.path.exists(src):
        return None
    _, ext = os.path.splitext(src)
    if ext.lower() == ".wav":
        try:
            with wave.open(src, "rb") as f:
                if f.getnchannels() > 0:
                    return src
        except wave.Error:
            pass
    out = str(INSTALL_DIR / f"tmp_conv_{int(time.time())}.wav")
    cmd = ["ffmpeg", "-y", "-v", "error", "-i", src,
           "-ar", str(SAMPLE_RATE), "-ac", "1", "-c:a", "pcm_s16le", out]
    try:
        subprocess.run(cmd, check=True, stdout=subprocess.DEVNULL, stderr=subprocess.PIPE)
        return out
    except (subprocess.CalledProcessError, FileNotFoundError):
        return None

def get_saved_voices():
    if not VOICES_DIR.exists():
        return []
    return sorted([f.stem for f in VOICES_DIR.glob("*.wav")])

# ========== サイドバー ==========
with st.sidebar:
    st.markdown("### ⚙️ 設定")

    model_size  = st.selectbox("🤖 モデルサイズ", ["1.7B（高品質）", "0.6B（軽量）"])
    size_key    = "1.7B" if "1.7B" in model_size else "0.6B"
    mode_label  = st.selectbox("🎛️ モード", [
        "Voice Cloning（音声クローン）",
        "Custom Voice（プリセット話者）",
        "Voice Design（声質デザイン）",
    ])
    mode_short  = mode_label.split("（")[0]

    st.markdown("---")
    st.markdown("### 📊 システム情報")
    st.caption(f"OS: {platform.system()} {platform.machine()}")

    backend_labels = {
        "mlx":  ("🍎 MLX (Apple Silicon)", "#27ae60"),
        "cuda": ("⚡ PyTorch CUDA",         "#2980b9"),
        "cpu":  ("🖥️ PyTorch CPU",          "#e67e22"),
        None:   ("❌ バックエンドなし",       "#e74c3c"),
    }
    label, color = backend_labels[BACKEND]
    st.markdown(f'<span class="env-badge" style="background:{color}">{label}</span>',
                unsafe_allow_html=True)
    if BACKEND is None:
        st.error("バックエンドが見つかりません")
        if IS_MAC:
            st.code("pip install mlx-audio", language="bash")
        else:
            st.code("pip install qwen-tts\npip install torch", language="bash")

    current_folder = MODEL_FOLDERS.get((size_key, mode_short), "")
    current_model  = get_model_path(current_folder)
    if current_model:
        st.success(f"✓ モデル検出\n`{current_folder}`")
    else:
        st.warning(f"⚠ モデル未検出\n`{current_folder}`")
        with st.expander("ダウンロード方法"):
            hf_name = current_folder.replace("-8bit", "").replace(
                "Qwen3-TTS-12Hz-", "Qwen/Qwen3-TTS-12Hz-")
            st.code(
                f"huggingface-cli download {hf_name} \\\n"
                f"  --local-dir ./models/{current_folder}",
                language="bash",
            )

    st.markdown("---")
    if IS_WIN and BACKEND == "cpu":
        st.warning("⚠ CPU モード: 処理に数分かかります")
    elif IS_WIN and BACKEND == "cuda":
        try:
            import torch
            st.caption(f"CUDA: {torch.version.cuda}")
            st.caption(f"GPU: {torch.cuda.get_device_name(0)}")
        except Exception:
            pass

# ========== メインレイアウト ==========
col_left, col_right = st.columns([1, 1], gap="large")

# ============================
# 左カラム：入力
# ============================
with col_left:

    if "Voice Cloning" in mode_label:
        st.markdown("#### 🎤 参照音声（Reference Audio）")
        st.caption("クローンしたい声のサンプルをアップロード（3〜5秒推奨）")

        clone_src = st.radio("音声の取得方法",
                             ["新しくアップロード", "保存済みの声を使う"], horizontal=True)

        ref_audio_path = None
        ref_text_val   = ""

        if clone_src == "新しくアップロード":
            uploaded = st.file_uploader("参照音声", type=["wav","mp3","m4a","flac"],
                                        label_visibility="collapsed")
            if uploaded:
                st.audio(uploaded)
                suffix = Path(uploaded.name).suffix
                tmp = tempfile.NamedTemporaryFile(delete=False, suffix=suffix,
                                                  dir=str(INSTALL_DIR))
                tmp.write(uploaded.read())
                tmp.close()
                ref_audio_path = tmp.name
                st.success(f"✓ {uploaded.name}")
            else:
                st.markdown("""
                <div class="hint-box">💡 WAV推奨・3〜5秒・雑音なし・16kHz以上</div>
                """, unsafe_allow_html=True)
        else:
            saved = get_saved_voices()
            if saved:
                sel = st.selectbox("保存済みの声", saved)
                ref_audio_path = str(VOICES_DIR / f"{sel}.wav")
                txt = VOICES_DIR / f"{sel}.txt"
                if txt.exists():
                    ref_text_val = txt.read_text(encoding="utf-8").strip()
                st.audio(ref_audio_path)
            else:
                st.info("保存済みの声がありません。")

        st.markdown("")
        st.markdown("#### 📝 参照テキスト（Reference Text）")
        ref_text_val = st.text_area("参照テキスト", value=ref_text_val,
            placeholder="例: こんにちは。これはクローニングのテストです。",
            height=80, label_visibility="collapsed")
        use_xvector = st.checkbox("🔬 x-vector のみ（参照テキスト不要・品質低下あり）")
        if use_xvector:
            ref_text_val = "."

        save_voice = st.checkbox("✅ この声を名前をつけて保存する")
        voice_name = st.text_input("保存名", placeholder="例: Boss, Mom") if save_voice else ""

    elif "Custom Voice" in mode_label:
        st.markdown("#### 👤 話者の選択")
        spk_map = {
            "Japanese": ["Ono_Anna"],
            "English":  ["Ryan","Aiden","Ethan","Chelsie","Serena","Vivian"],
            "Chinese":  ["Vivian","Serena","Uncle_Fu","Dylan","Eric"],
            "Korean":   ["Sohee"],
        }
        lang     = st.selectbox("言語", list(spk_map.keys()))
        speaker  = st.selectbox("話者", spk_map[lang])
        st.markdown("#### 🎭 感情・口調")
        emotions = ["Normal tone","Sad and crying, speaking slowly",
                    "Excited and happy, speaking very fast",
                    "Angry and shouting","Whispering quietly"]
        preset   = st.selectbox("プリセット", emotions)
        custom_e = st.text_input("カスタム指定（英語）", placeholder="例: Calm and professional")
        instruct_val = custom_e.strip() or preset
        st.markdown("#### ⚡ 速度")
        speed_opt = st.select_slider("速度", [0.8,1.0,1.3], value=1.0,
            format_func=lambda x:{0.8:"遅(0.8x)",1.0:"普通(1.0x)",1.3:"速(1.3x)"}[x])
    else:
        st.markdown("#### 🎨 声質のデザイン")
        instruct_val = st.text_area("声質の説明（英語）",
            placeholder="例: A calm middle-aged Japanese male voice, speaking slowly.",
            height=100, label_visibility="collapsed")

    st.markdown("---")
    st.markdown("#### 💬 生成テキスト（Target Text）")
    target_text = st.text_area("生成テキスト",
        placeholder="例: 今日も一日、お疲れ様でした。",
        height=110, label_visibility="collapsed")

    # トーン調整（Voice Cloning モードのみ）
    tone_instruct = ""
    if "Voice Cloning" in mode_label:
        st.markdown("#### 🎭 トーン調整")
        tone_presets = {
            "通常（指定なし）":         "",
            "ゆっくり丁寧に":           "Speak slowly and politely.",
            "明るく元気に":             "Speak in a bright and energetic tone.",
            "落ち着いてナレーション風": "Speak calmly like a narrator.",
            "ニュースキャスター風":     "Speak clearly and professionally like a news anchor.",
            "悲しそうに・しんみりと":   "Speak sadly and solemnly.",
        }
        tone_label  = st.selectbox("話し方のスタイル", list(tone_presets.keys()),
                                   label_visibility="collapsed")
        tone_custom = st.text_input("カスタム指定（英語）",
                                    placeholder="例: Speak with excitement and enthusiasm.")
        tone_instruct = tone_custom.strip() or tone_presets[tone_label]

    generate_btn = st.button("🎵  Clone & Generate", use_container_width=True)

# ============================
# 右カラム：出力
# ============================
with col_right:
    st.markdown("#### 🔊 生成音声（Generated Audio）")
    audio_ph  = st.empty()
    status_ph = st.empty()
    log_ph    = st.empty()

    with audio_ph.container():
        st.markdown("""
        <div style="background:#f0f2f6;border:2px dashed #ccc;border-radius:10px;
                    padding:3rem;text-align:center;color:#999;">
            <div style="font-size:3rem;">🎧</div>
            <div>生成後にここに音声が表示されます</div>
        </div>""", unsafe_allow_html=True)

    with status_ph.container():
        st.markdown('<div class="status-bar">> 待機中...</div>',
                    unsafe_allow_html=True)

    st.markdown("---")
    with st.expander("📖 使い方", expanded=True):
        st.markdown("""
**Voice Cloning:** 参照音声をアップロード → 参照テキスト入力 → 生成テキスト入力 → Generate

**Custom Voice:** 話者・感情を選んで生成テキスト入力

**Voice Design:** 声の特徴を英語で記述して生成
        """)

# ========== 生成処理 ==========
def upd(msg):
    status_ph.markdown(f'<div class="status-bar">> {msg}</div>',
                       unsafe_allow_html=True)

if generate_btn:
    errs = []
    if not target_text.strip():
        errs.append("生成テキストを入力してください。")
    if "Voice Cloning" in mode_label and not ref_audio_path:
        errs.append("参照音声をアップロードしてください。")
    if BACKEND is None:
        errs.append("バックエンドがインストールされていません。")
    if not current_model:
        errs.append(f"モデルが見つかりません: `{current_folder}`")
    for e in errs:
        st.error(f"❌ {e}")

    if not errs:
        OUTPUT_DIR.mkdir(parents=True, exist_ok=True)
        ts        = datetime.now().strftime("%H-%M-%S")
        clean_t   = re.sub(r'[^\w\s-]', '', target_text)[:20].strip().replace(' ','_') or "audio"
        out_name  = f"{ts}_{clean_t}.wav"
        temp_dir  = str(INSTALL_DIR / f"tmp_{int(time.time())}")
        os.makedirs(temp_dir, exist_ok=True)

        try:
            # ===== MLX バックエンド（Mac Apple Silicon）=====
            if BACKEND == "mlx":
                from mlx_audio.tts.utils import load_model
                from mlx_audio.tts.generate import generate_audio

                cache_key = f"mlx_model_{current_folder}"
                if cache_key not in st.session_state:
                    upd(f"モデルを読み込んでいます（MLX）: {current_folder} ...")
                    st.session_state[cache_key] = load_model(current_model)
                model = st.session_state[cache_key]
                upd("音声を生成しています（MLX）...")

                if "Voice Cloning" in mode_label:
                    wav_path = convert_to_wav(ref_audio_path)
                    if not wav_path:
                        st.error("❌ 参照音声の変換失敗。ffmpegを確認してください。")
                        st.stop()
                    gen_kwargs = dict(
                        model=model, text=target_text,
                        ref_audio=wav_path, ref_text=ref_text_val,
                        output_path=temp_dir,
                    )
                    if tone_instruct:
                        gen_kwargs["instruct"] = tone_instruct
                    generate_audio(**gen_kwargs)
                    if save_voice and voice_name.strip():
                        VOICES_DIR.mkdir(exist_ok=True)
                        safe = re.sub(r'[^\w-]','',voice_name).strip()
                        shutil.copy(wav_path, str(VOICES_DIR/f"{safe}.wav"))
                        if ref_text_val:
                            (VOICES_DIR/f"{safe}.txt").write_text(ref_text_val, encoding="utf-8")
                    if wav_path != ref_audio_path and os.path.exists(wav_path):
                        os.remove(wav_path)
                elif "Custom Voice" in mode_label:
                    generate_audio(model=model, text=target_text,
                                   voice=speaker, instruct=instruct_val,
                                   speed=speed_opt, output_path=temp_dir)
                else:
                    generate_audio(model=model, text=target_text,
                                   instruct=instruct_val, output_path=temp_dir)

                src_wav = os.path.join(temp_dir, "audio_000.wav")

            # ===== PyTorch バックエンド（Windows CUDA / CPU）=====
            else:
                import torch
                import soundfile as sf
                from qwen_tts import Qwen3TTSModel

                device = "cuda" if BACKEND == "cuda" else "cpu"
                dtype  = torch.float16 if BACKEND == "cuda" else torch.float32

                cache_key = f"pt_model_{current_folder}"
                if cache_key not in st.session_state:
                    upd(f"モデルを読み込んでいます（{device.upper()}）: {current_folder} ...")
                    st.session_state[cache_key] = Qwen3TTSModel.from_pretrained(
                        current_model,
                        device_map=device,
                        dtype=dtype,
                        attn_implementation="sdpa",
                    )
                model = st.session_state[cache_key]
                upd(f"音声を生成しています（{device.upper()}）...")

                src_wav = os.path.join(temp_dir, "output.wav")

                if "Voice Cloning" in mode_label:
                    clone_kwargs = dict(
                        text=target_text,
                        ref_audio=ref_audio_path,
                        ref_text=ref_text_val if not use_xvector else None,
                    )
                    if tone_instruct:
                        clone_kwargs["instruct"] = tone_instruct
                    wavs, sr = model.generate_voice_clone(**clone_kwargs)
                    sf.write(src_wav, wavs[0], sr)
                    if save_voice and voice_name.strip():
                        VOICES_DIR.mkdir(exist_ok=True)
                        safe = re.sub(r'[^\w-]','',voice_name).strip()
                        shutil.copy(ref_audio_path, str(VOICES_DIR/f"{safe}.wav"))
                        if ref_text_val:
                            (VOICES_DIR/f"{safe}.txt").write_text(ref_text_val, encoding="utf-8")
                else:
                    # Custom Voice / Voice Design
                    wavs, sr = model.generate(text=target_text)
                    sf.write(src_wav, wavs[0], sr)

            # 出力先に移動
            final_dir = OUTPUT_DIR / mode_short.replace(" ","")
            final_dir.mkdir(parents=True, exist_ok=True)
            final_path = str(final_dir / out_name)
            if os.path.exists(src_wav):
                shutil.move(src_wav, final_path)
            shutil.rmtree(temp_dir, ignore_errors=True)

            upd(f"✓ 完了 → {final_path}")

            with audio_ph.container():
                st.success("✅ 生成完了！")
                with open(final_path, "rb") as f:
                    ab = f.read()
                st.audio(ab, format="audio/wav")
                st.download_button("⬇️ WAVをダウンロード", ab,
                                   file_name=out_name, mime="audio/wav",
                                   use_container_width=True)

            with log_ph.container():
                with st.expander("📋 生成パラメータ"):
                    p = {"モード": mode_label, "バックエンド": BACKEND,
                         "モデル": current_folder, "生成テキスト": target_text}
                    if "Voice Cloning" in mode_label:
                        p["参照テキスト"] = ref_text_val
                    elif "Custom Voice" in mode_label:
                        p["話者"] = speaker; p["感情"] = instruct_val; p["速度"] = speed_opt
                    else:
                        p["声質指定"] = instruct_val
                    st.json(p)

        except Exception as e:
            upd(f"❌ エラー: {e}")
            st.error(f"```\n{e}\n```")

st.markdown("---")
st.markdown("""
<div style="text-align:center;color:#888;font-size:.8rem">
    Qwen3-TTS ローカルGUI | Powered by Qwen Team (Alibaba Cloud) |
    <a href="https://github.com/QwenLM/Qwen3-TTS">GitHub</a>
</div>""", unsafe_allow_html=True)