完善fastapi接口

api/schemas/frame.py

@@ -0,0 +1,37 @@
+"""
+Frame/Template rendering API schemas
+"""
+
+from typing import Optional
+from pydantic import BaseModel, Field
+
+
+class FrameRenderRequest(BaseModel):
+    """Frame rendering request"""
+    template: str = Field(
+        ..., 
+        description="Template key (e.g., '1080x1920/default.html'). Can also be just filename (e.g., 'default.html') to use default size."
+    )
+    title: Optional[str] = Field(None, description="Frame title (optional)")
+    text: str = Field(..., description="Frame text content")
+    image: str = Field(..., description="Image path or URL")
+    
+    class Config:
+        json_schema_extra = {
+            "example": {
+                "template": "1080x1920/default.html",
+                "title": "Sample Title",
+                "text": "This is a sample text for the frame.",
+                "image": "resources/example.png"
+            }
+        }
+
+
+class FrameRenderResponse(BaseModel):
+    """Frame rendering response"""
+    success: bool = True
+    message: str = "Success"
+    frame_path: str = Field(..., description="Path to generated frame image")
+    width: int = Field(..., description="Frame width in pixels")
+    height: int = Field(..., description="Frame height in pixels")
+

api/schemas/resources.py

@@ -0,0 +1,57 @@
+"""
+Resource discovery API schemas
+"""
+
+from typing import List, Optional
+from pydantic import BaseModel, Field
+
+
+class WorkflowInfo(BaseModel):
+    """Workflow information"""
+    name: str = Field(..., description="Workflow filename")
+    display_name: str = Field(..., description="Display name with source info")
+    source: str = Field(..., description="Source (runninghub or selfhost)")
+    path: str = Field(..., description="Full path to workflow file")
+    key: str = Field(..., description="Workflow key (source/name)")
+    workflow_id: Optional[str] = Field(None, description="RunningHub workflow ID (if applicable)")
+
+
+class WorkflowListResponse(BaseModel):
+    """Workflow list response"""
+    success: bool = True
+    message: str = "Success"
+    workflows: List[WorkflowInfo] = Field(..., description="List of available workflows")
+
+
+class TemplateInfo(BaseModel):
+    """Template information"""
+    name: str = Field(..., description="Template filename")
+    display_name: str = Field(..., description="Display name")
+    size: str = Field(..., description="Size (e.g., 1080x1920)")
+    width: int = Field(..., description="Width in pixels")
+    height: int = Field(..., description="Height in pixels")
+    orientation: str = Field(..., description="Orientation (portrait/landscape/square)")
+    path: str = Field(..., description="Full path to template file")
+    key: str = Field(..., description="Template key (size/name)")
+
+
+class TemplateListResponse(BaseModel):
+    """Template list response"""
+    success: bool = True
+    message: str = "Success"
+    templates: List[TemplateInfo] = Field(..., description="List of available templates")
+
+
+class BGMInfo(BaseModel):
+    """BGM information"""
+    name: str = Field(..., description="BGM filename")
+    path: str = Field(..., description="Full path to BGM file")
+    source: str = Field(..., description="Source (default or custom)")
+
+
+class BGMListResponse(BaseModel):
+    """BGM list response"""
+    success: bool = True
+    message: str = "Success"
+    bgm_files: List[BGMInfo] = Field(..., description="List of available BGM files")
+

api/schemas/tts.py

@@ -2,19 +2,32 @@
 TTS API schemas
 """
 
+from typing import Optional
 from pydantic import BaseModel, Field
 
 
 class TTSSynthesizeRequest(BaseModel):
     """TTS synthesis request"""
     text: str = Field(..., description="Text to synthesize")
-    voice_id: str = Field("[Chinese] zh-CN Yunjian", description="Voice ID")
+    workflow: Optional[str] = Field(
+        None, 
+        description="TTS workflow key (e.g., 'runninghub/tts_edge.json' or 'selfhost/tts_edge.json'). If not specified, uses default workflow from config."
+    )
+    ref_audio: Optional[str] = Field(
+        None, 
+        description="Reference audio path for voice cloning (optional). Can be a local file path or URL."
+    )
+    voice_id: Optional[str] = Field(
+        None, 
+        description="Voice ID (deprecated, use workflow instead)"
+    )
     
     class Config:
         json_schema_extra = {
             "example": {
                 "text": "Hello, welcome to Pixelle-Video!",
-                "voice_id": "[Chinese] zh-CN Yunjian"
+                "workflow": "runninghub/tts_edge.json",
+                "ref_audio": None
             }
         }
 

api/schemas/video.py

@@ -23,7 +23,20 @@ class VideoGenerateRequest(BaseModel):
     
     # === Basic Config ===
     n_scenes: int = Field(5, ge=1, le=20, description="Number of scenes (generate mode only)")
-    voice_id: str = Field("[Chinese] zh-CN Yunjian", description="TTS voice ID")
+    
+    # === TTS Parameters ===
+    tts_workflow: Optional[str] = Field(
+        None, 
+        description="TTS workflow key (e.g., 'runninghub/tts_edge.json'). If not specified, uses default workflow from config."
+    )
+    ref_audio: Optional[str] = Field(
+        None, 
+        description="Reference audio path for voice cloning (optional)"
+    )
+    voice_id: Optional[str] = Field(
+        None, 
+        description="(Deprecated) TTS voice ID for legacy compatibility"
+    )
     
     # === LLM Parameters ===
     min_narration_words: int = Field(5, ge=1, le=100, description="Min narration words")