feat: add ASR provider support for voice co-creation

backend/app/services/provider_policy.py

@@ -4,7 +4,7 @@ from dataclasses import dataclass
 from enum import Enum
 from typing import Literal, Protocol, TypeAlias
 
-ProviderType: TypeAlias = Literal["text", "image", "tts", "storybook"]
+ProviderType: TypeAlias = Literal["text", "image", "tts", "storybook", "asr"]
 
 
 class RoutingStrategy(str, Enum):
@@ -36,6 +36,7 @@ class ProviderSettings(Protocol):
     image_providers: list[str]
     tts_providers: list[str]
     storybook_providers: list[str]
+    asr_providers: list[str]
     enable_demo_providers: bool
 
 
@@ -71,6 +72,14 @@ CAPABILITY_POLICIES: dict[ProviderType, CapabilityPolicy] = {
         default_providers=("storybook_primary",),
         demo_provider="demo",
     ),
+    "asr": CapabilityPolicy(
+        capability="asr",
+        label="语音识别",
+        description="将孩子上传的语音回合转写为文本输入。",
+        settings_attr="asr_providers",
+        default_providers=("demo",),
+        demo_provider="demo",
+    ),
 }
 
 
@@ -94,6 +103,8 @@ API_KEY_MAP: dict[str, str] = {
     "antigravity_api_key": "antigravity_api_key",
     "image_primary": "image_api_key",
     "image_api_key": "image_api_key",
+    # ASR
+    "openai_asr": "openai_api_key",
     # TTS
     "minimax": "minimax_api_key",
     "minimax_api_key": "minimax_api_key",