auto-save 2026-05-18 00:29 (~8)

api/.env.example

@@ -18,15 +18,29 @@ LOCAL_ASR_MODEL=mlx-community/whisper-tiny
 LOCAL_ASR_TIMEOUT_SECONDS=180
 TRANSLATE_MODEL=gemini-2.5-flash
 REWRITE_MODEL=gemini-2.5-pro
-IMAGE_MODEL=gemini-3-pro-image-preview
+IMAGE_BASE_URL=https://ai.skg.com/ezlink/v1
+IMAGE_API_KEY=
+IMAGE_MODEL=gpt-image-2
+GPT_IMAGE_MODEL=gpt-image-2
+SUBJECT_ASSET_IMAGE_MODEL=gpt-image-2
+SUBJECT_ASSET_IMAGE_MODELS=gpt-image-2,gpt-image-1.5
 VIDEO_MODEL=seedance
 VIDEO_MODEL_SEEDANCE=seedance-2-fast
 VIDEO_MODEL_KLING=kling-omni
 VIDEO_MODEL_VEO3=veo-3.1-fast
 
-# 音频文案改写 + MiniMax 配音
+# 音频文案改写 + Azure OpenAI 配音
 AUDIO_REWRITE_MODEL=gemini-2.5-pro
 AUDIO_PRODUCT_BRIEF="SKG 智能按摩产品，主打日常肩颈、腰背、眼部、膝盖或足部放松；广告表达要高级、干净、可信，不做医疗疗效承诺。"
+VOICE_PROVIDER=azure_openai
+AZURE_OPENAI_BASE_URL=https://ai.skg.com/azure
+AZURE_OPENAI_API_KEY=
+AZURE_TTS_MODEL=gpt-4o-mini-tts
+AZURE_TTS_VOICE_ID=alloy
+AZURE_TTS_VOICE_POOL=alloy,verse,shimmer
+AZURE_TTS_PATH=/audio/speech
+
+# MiniMax 旧配音通道，保留兼容；默认不走
 MINIMAX_API_KEY=
 MINIMAX_TTS_BASE_URL=https://api.minimax.io
 MINIMAX_TTS_MODEL=speech-2.8-turbo

api/main.py

@@ -624,6 +624,17 @@ def video_uses_ark() -> bool:
     return "ark.cn-beijing.volces.com" in base or "ai.skg.com/doubao" in base
 
 
+def video_provider_name() -> str:
+    base = video_api_base()
+    if video_uses_poe():
+        return "poe"
+    if "ai.skg.com/doubao" in base:
+        return "doubao"
+    if "ark.cn-beijing.volces.com" in base:
+        return "ark"
+    return "custom"
+
+
 def video_api_base() -> str:
     if VIDEO_API_BASE_URL:
         return VIDEO_API_BASE_URL.rstrip("/")
@@ -2853,7 +2864,7 @@ def health() -> dict:
             "vision": VISION_MODEL,
             "image": IMAGE_MODEL,
             "image_base_url": IMAGE_BASE_URL or LLM_BASE_URL or "openai-default",
-            "image_fallbacks": [IMAGE_MODEL, GPT_IMAGE_MODEL, "gpt-image-1.5"],
+            "image_fallbacks": list(dict.fromkeys([IMAGE_MODEL, GPT_IMAGE_MODEL, "gpt-image-1.5"])),
             "subject_image": SUBJECT_ASSET_IMAGE_MODEL,
             "subject_image_fallbacks": SUBJECT_ASSET_IMAGE_MODELS,
             "voice_provider": VOICE_PROVIDER,
@@ -2868,7 +2879,7 @@ def health() -> dict:
             "minimax_configured": bool(MINIMAX_API_KEY),
             "video": VIDEO_MODEL,
             "video_aliases": VIDEO_MODEL_ALIASES,
-            "video_provider": "poe" if video_uses_poe() else ("ark" if video_uses_ark() else "custom"),
+            "video_provider": video_provider_name(),
             "video_base_url": video_api_base(),
             "video_configured": bool(video_api_key()),
             "video_create_paths": VIDEO_CREATE_PATHS,