D144 · 口型同步三路由上线: Wav2Lip→火山云端→Edge-TTS

- lipsync-adapter.py: 新增云端路由，Wav2Lip不可用时自动尝试火山改口型API
- volce-lipsync-adapter.py: 火山引擎口型同步适配器（STUB），需开通智能视觉服务
- 当前状态: Wav2Lip需CUDA GPU(本地不可用)，火山需AK/SK配置，TTS配音可用

铸渊 ICE-GL-ZY001 · D144 · 2026-06-24

video-ai-system/engines/lipsync-adapter.py

@@ -31,12 +31,13 @@ sys.path.insert(0, str(PROJECT_ROOT / "engines"))
 
 
 class LipSyncAdapter:
-    """口型适配器"""
+    """口型适配器 — 支持多路由: Wav2Lip(本地GPU) → 火山改口型(云端) → 回退"""
 
     def __init__(self, wav2lip_path=None, allow_fallback=False):
         self.wav2lip_path = wav2lip_path or PROJECT_ROOT / "tools" / "Wav2Lip"
         self.allow_fallback = allow_fallback
         self.available = self._check_wav2lip()
+        self._cloud_available = None
 
     def _check_wav2lip(self):
         """检查 Wav2Lip 是否可用"""
@@ -93,12 +94,27 @@ class LipSyncAdapter:
         # 确保输出目录存在
         output_path.parent.mkdir(parents=True, exist_ok=True)
 
-        # 方法1: Wav2Lip
+        # 方法1: Wav2Lip (本地GPU)
         if self.available:
             print(f"  🔧 使用 Wav2Lip...")
             result = self._run_wav2lip(video_path, audio_path, output_path)
             return result
 
+        # 方法2: 火山引擎云端改口型
+        cloud = self._check_cloud()
+        if cloud["available"]:
+            print(f"  ☁️  尝试火山改口型...")
+            result = self._run_cloud_lipsync(video_path, audio_path, output_path)
+            if result.get("success"):
+                return result
+            print(f"  ⚠️  云端改口型失败: {result.get('error')}")
+
+        # 方法3: Edge-TTS 配音替代（角色不说话时可用）
+        print(f"  💡 当前可用路由:")
+        print(f"     Route1 Wav2Lip:    {'✅' if self.available else '❌ 需要CUDA GPU'}")
+        print(f"     Route2 火山改口型: {'STUB 需开通' if not cloud.get('available') else '可用'}")
+        print(f"     Route3 Edge-TTS:   ✅ (voice-emotion-compiler已通过)")
+
         fallback_enabled = self.allow_fallback if allow_fallback is None else allow_fallback
         if not fallback_enabled:
             print(f"  ❌ Wav2Lip 不可用，口型同步未执行")
@@ -124,6 +140,37 @@ class LipSyncAdapter:
             "warning": "Wav2Lip 不可用，口型未同步。该文件只能用于调试占位。"
         }
 
+    def _check_cloud(self):
+        """检查云端改口型服务是否可用"""
+        if self._cloud_available is not None:
+            return self._cloud_available
+        try:
+            import importlib.util
+            spec = importlib.util.spec_from_file_location(
+                'volce',
+                str(PROJECT_ROOT / 'engines' / 'volce-lipsync-adapter.py')
+            )
+            mod = importlib.util.module_from_spec(spec)
+            spec.loader.exec_module(mod)
+            self._cloud_available = mod.check_available()
+        except Exception as e:
+            self._cloud_available = {"available": False, "reason": f"云端适配器不可用: {e}"}
+        return self._cloud_available
+
+    def _run_cloud_lipsync(self, video_path, audio_path, output_path):
+        """通过火山引擎云端改口型"""
+        try:
+            import importlib.util
+            spec = importlib.util.spec_from_file_location(
+                'volce',
+                str(PROJECT_ROOT / 'engines' / 'volce-lipsync-adapter.py')
+            )
+            mod = importlib.util.module_from_spec(spec)
+            spec.loader.exec_module(mod)
+            return mod.submit_lipsync(video_path, audio_path)
+        except Exception as e:
+            return {"success": False, "error": f"云端适配器调用失败: {e}", "method": "cloud-stub"}
+
     def _run_wav2lip(self, video_path, audio_path, output_path):
         """运行 Wav2Lip"""
         import subprocess

video-ai-system/engines/volce-lipsync-adapter.py

@@ -0,0 +1,104 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+volce-lipsync-adapter.py · 火山引擎口型同步适配器
+D144 · 铸渊 ICE-GL-ZY001
+
+对接火山引擎智能视觉服务"视频改口型"API。
+需要预先开通服务并获取 AK/SK。
+
+文档:
+  https://www.volcengine.com/docs/85128/1462622
+
+依赖:
+  pip install volcengine-python-sdk
+
+当前状态: STUB — 需要开通服务并配置 AK/SK 后才能激活
+"""
+
+import os
+import json
+import base64
+import hashlib
+import hmac
+import urllib.request
+import urllib.parse
+from datetime import datetime, timezone
+
+
+SECRETS_FILE = "/Users/bingshuolingdianyuanhe/Documents/guanghulab-local-secrets/video-ai-system.env"
+
+
+def _load_secrets():
+    """加载本地密钥文件"""
+    secrets = {}
+    if os.path.isfile(SECRETS_FILE):
+        with open(SECRETS_FILE) as f:
+            for line in f:
+                line = line.strip()
+                if line and not line.startswith("#") and "=" in line:
+                    k, v = line.split("=", 1)
+                    secrets[k.strip()] = v.strip().strip('"').strip("'")
+    return secrets
+
+
+def check_available():
+    """检查火山口型同步服务是否可用"""
+    secrets = _load_secrets()
+    ak = secrets.get("VOLCE_ACCESS_KEY", "")
+    sk = secrets.get("VOLCE_SECRET_KEY", "")
+
+    if not ak or not sk:
+        return {
+            "available": False,
+            "reason": "未配置 VOLCE_ACCESS_KEY / VOLCE_SECRET_KEY",
+            "help": "请前往火山引擎控制台 → 访问控制 → 创建Access Key → 写入本地密钥文件"
+        }
+
+    return {
+        "available": True,
+        "method": "volce-video-lipsync"
+    }
+
+
+def submit_lipsync(video_path, audio_path, source_lang="zh", target_lang="zh"):
+    """
+    提交视频改口型任务
+
+    注意: 当前为 STUB 实现。火山引擎视频改口型 API 需要:
+    1. 开通智能视觉服务
+    2. 获取独立 AK/SK（不是 JIMENG_API_KEY）
+    3. 上传视频到火山 TOS 对象存储
+    4. 提交异步任务并轮询结果
+
+    :param video_path: 输入视频
+    :param audio_path: 对白音频
+    :param source_lang: 源语言
+    :param target_lang: 目标语言
+    :return: {success, output_path, method}
+    """
+    status = check_available()
+    if not status["available"]:
+        return {
+            "success": False,
+            "error": status["reason"],
+            "method": "volce-stub"
+        }
+
+    # STUB: 需要完整开通后才能调用
+    return {
+        "success": False,
+        "error": "火山视频改口型 API 尚未完整接入（STUB）",
+        "method": "volce-stub",
+        "next_steps": [
+            "开通火山引擎智能视觉服务",
+            "配置 VOLCE_ACCESS_KEY / VOLCE_SECRET_KEY",
+            "实现视频上传到 TOS 的逻辑",
+            "实现异步任务提交和轮询"
+        ]
+    }
+
+
+if __name__ == "__main__":
+    result = check_available()
+    print(json.dumps(result, ensure_ascii=False, indent=2))