wx44wx/AI-VideoAssistant
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Use openai compatible as vendor

Browse Source
This commit is contained in:
Xin Wang
2026-02-12 18:44:55 +08:00
parent 260ff621bf
commit ff3a03b1ad
23 changed files with 822 additions and 905 deletions
Download Patch File Download Diff File Expand all files Collapse all files

20
api/app/routers/asr.py
View File

@@ -16,16 +16,22 @@ from ..schemas import (
router = APIRouter(prefix="/asr", tags=["ASR Models"])
SILICONFLOW_DEFAULT_ASR_MODEL = "FunAudioLLM/SenseVoiceSmall"
OPENAI_COMPATIBLE_DEFAULT_ASR_MODEL = "FunAudioLLM/SenseVoiceSmall"
def _is_siliconflow_vendor(vendor: str) -> bool:
return (vendor or "").strip().lower() in {"siliconflow", "硅基流动"}
def _is_openai_compatible_vendor(vendor: str) -> bool:
normalized = (vendor or "").strip().lower()
return normalized in {
"openai compatible",
"openai-compatible",
"siliconflow", # backward compatibility
"硅基流动", # backward compatibility
}
def _default_asr_model(vendor: str) -> str:
if _is_siliconflow_vendor(vendor):
return SILICONFLOW_DEFAULT_ASR_MODEL
if _is_openai_compatible_vendor(vendor):
return OPENAI_COMPATIBLE_DEFAULT_ASR_MODEL
return "whisper-1"
@@ -129,7 +135,7 @@ def test_asr_model(
# 连接性测试优先，避免依赖真实音频输入
headers = {"Authorization": f"Bearer {model.api_key}"}
with httpx.Client(timeout=60.0) as client:
if model.vendor.lower() in ["siliconflow", "paraformer"]:
if _is_openai_compatible_vendor(model.vendor) or model.vendor.lower() == "paraformer":
response = client.get(f"{model.base_url}/asr", headers=headers)
elif model.vendor.lower() == "openai":
response = client.get(f"{model.base_url}/audio/models", headers=headers)
@@ -258,7 +264,7 @@ async def preview_asr_model(
raise HTTPException(status_code=400, detail="Uploaded audio file is empty")
effective_api_key = (api_key or "").strip() or (model.api_key or "").strip()
if not effective_api_key and _is_siliconflow_vendor(model.vendor):
if not effective_api_key and _is_openai_compatible_vendor(model.vendor):
effective_api_key = os.getenv("SILICONFLOW_API_KEY", "").strip()
if not effective_api_key:
raise HTTPException(status_code=400, detail=f"API key is required for ASR model: {model.name}")

17
api/app/routers/assistants.py
View File

@@ -13,8 +13,13 @@ from ..schemas import (
router = APIRouter(prefix="/assistants", tags=["Assistants"])
def _is_siliconflow_vendor(vendor: Optional[str]) -> bool:
return (vendor or "").strip().lower() in {"siliconflow", "硅基流动"}
def _is_openai_compatible_vendor(vendor: Optional[str]) -> bool:
return (vendor or "").strip().lower() in {
"siliconflow",
"硅基流动",
"openai compatible",
"openai-compatible",
}
def _resolve_runtime_metadata(db: Session, assistant: Assistant) -> dict:
@@ -47,11 +52,11 @@ def _resolve_runtime_metadata(db: Session, assistant: Assistant) -> dict:
if assistant.asr_model_id:
asr = db.query(ASRModel).filter(ASRModel.id == assistant.asr_model_id).first()
if asr:
asr_provider = "siliconflow" if _is_siliconflow_vendor(asr.vendor) else "buffered"
asr_provider = "openai_compatible" if _is_openai_compatible_vendor(asr.vendor) else "buffered"
metadata["services"]["asr"] = {
"provider": asr_provider,
"model": asr.model_name or asr.name,
"apiKey": asr.api_key if asr_provider == "siliconflow" else None,
"apiKey": asr.api_key if asr_provider == "openai_compatible" else None,
}
else:
warnings.append(f"ASR model not found: {assistant.asr_model_id}")
@@ -61,12 +66,12 @@ def _resolve_runtime_metadata(db: Session, assistant: Assistant) -> dict:
elif assistant.voice:
voice = db.query(Voice).filter(Voice.id == assistant.voice).first()
if voice:
tts_provider = "siliconflow" if _is_siliconflow_vendor(voice.vendor) else "edge"
tts_provider = "openai_compatible" if _is_openai_compatible_vendor(voice.vendor) else "edge"
metadata["services"]["tts"] = {
"enabled": True,
"provider": tts_provider,
"model": voice.model,
"apiKey": voice.api_key if tts_provider == "siliconflow" else None,
"apiKey": voice.api_key if tts_provider == "openai_compatible" else None,
"voice": voice.voice_key or voice.id,
"speed": assistant.speed or voice.speed,
}

8
api/app/routers/tools.py
View File

@@ -467,7 +467,13 @@ def _test_asr_model(db: Session, model_id: str, result: AutotestResult):
headers = {"Authorization": f"Bearer {model.api_key}"}
with httpx.Client(timeout=30.0) as client:
if model.vendor.lower() in ["siliconflow", "paraformer"]:
normalized_vendor = (model.vendor or "").strip().lower()
if normalized_vendor in [
"openai compatible",
"openai-compatible",
"siliconflow", # backward compatibility
"paraformer",
]:
response = client.get(
f"{model.base_url}/asr",
headers=headers

32
api/app/routers/voices.py
View File

@@ -13,20 +13,26 @@ from ..schemas import VoiceCreate, VoiceOut, VoicePreviewRequest, VoicePreviewRe
router = APIRouter(prefix="/voices", tags=["Voices"])
SILICONFLOW_DEFAULT_MODEL = "FunAudioLLM/CosyVoice2-0.5B"
OPENAI_COMPATIBLE_DEFAULT_MODEL = "FunAudioLLM/CosyVoice2-0.5B"
def _is_siliconflow_vendor(vendor: str) -> bool:
return vendor.strip().lower() in {"siliconflow", "硅基流动"}
def _is_openai_compatible_vendor(vendor: str) -> bool:
normalized = (vendor or "").strip().lower()
return normalized in {
"openai compatible",
"openai-compatible",
"siliconflow", # backward compatibility
"硅基流动", # backward compatibility
}
def _default_base_url(vendor: str) -> Optional[str]:
if _is_siliconflow_vendor(vendor):
if _is_openai_compatible_vendor(vendor):
return "https://api.siliconflow.cn/v1"
return None
def _build_siliconflow_voice_key(voice: Voice, model: str) -> str:
def _build_openai_compatible_voice_key(voice: Voice, model: str) -> str:
if voice.voice_key:
return voice.voice_key
if ":" in voice.id:
@@ -65,8 +71,8 @@ def create_voice(data: VoiceCreate, db: Session = Depends(get_db)):
model = data.model
voice_key = data.voice_key
if _is_siliconflow_vendor(vendor):
model = model or SILICONFLOW_DEFAULT_MODEL
if _is_openai_compatible_vendor(vendor):
model = model or OPENAI_COMPATIBLE_DEFAULT_MODEL
if not voice_key:
raw_id = (data.id or data.name).strip()
voice_key = raw_id if ":" in raw_id else f"{model}:{raw_id}"
@@ -115,11 +121,11 @@ def update_voice(id: str, data: VoiceUpdate, db: Session = Depends(get_db)):
update_data["vendor"] = update_data["vendor"].strip()
vendor_for_defaults = update_data.get("vendor", voice.vendor)
if _is_siliconflow_vendor(vendor_for_defaults):
model = update_data.get("model") or voice.model or SILICONFLOW_DEFAULT_MODEL
if _is_openai_compatible_vendor(vendor_for_defaults):
model = update_data.get("model") or voice.model or OPENAI_COMPATIBLE_DEFAULT_MODEL
voice_key = update_data.get("voice_key") or voice.voice_key
update_data["model"] = model
update_data["voice_key"] = voice_key or _build_siliconflow_voice_key(voice, model)
update_data["voice_key"] = voice_key or _build_openai_compatible_voice_key(voice, model)
for field, value in update_data.items():
setattr(voice, field, value)
@@ -152,7 +158,7 @@ def preview_voice(id: str, data: VoicePreviewRequest, db: Session = Depends(get_
raise HTTPException(status_code=400, detail="Preview text cannot be empty")
api_key = (data.api_key or "").strip() or (voice.api_key or "").strip()
if not api_key and _is_siliconflow_vendor(voice.vendor):
if not api_key and _is_openai_compatible_vendor(voice.vendor):
api_key = os.getenv("SILICONFLOW_API_KEY", "").strip()
if not api_key:
raise HTTPException(status_code=400, detail=f"API key is required for voice: {voice.name}")
@@ -161,11 +167,11 @@ def preview_voice(id: str, data: VoicePreviewRequest, db: Session = Depends(get_
if not base_url:
raise HTTPException(status_code=400, detail=f"Base URL is required for voice: {voice.name}")
model = voice.model or SILICONFLOW_DEFAULT_MODEL
model = voice.model or OPENAI_COMPATIBLE_DEFAULT_MODEL
payload = {
"model": model,
"input": text,
"voice": voice.voice_key or _build_siliconflow_voice_key(voice, model),
"voice": voice.voice_key or _build_openai_compatible_voice_key(voice, model),
"response_format": "mp3",
"speed": data.speed if data.speed is not None else voice.speed,
}

20
api/docs/asr.md
View File

@@ -20,7 +20,7 @@ interface ASRModel {
id: string; // 模型唯一标识 (8位UUID)
user_id: number; // 所属用户ID
name: string; // 模型显示名称
vendor: string; // 供应商: "OpenAI" | "SiliconFlow" | "Paraformer" | 等
vendor: string; // 供应商: "OpenAI Compatible" | "Paraformer" | 等
language: string; // 识别语言: "zh" | "en" | "Multi-lingual"
base_url: string; // API Base URL
api_key: string; // API Key
@@ -64,7 +64,7 @@ GET /api/v1/asr
"id": "abc12345",
"user_id": 1,
"name": "Whisper 多语种识别",
"vendor": "OpenAI",
"vendor": "OpenAI Compatible",
"language": "Multi-lingual",
"base_url": "https://api.openai.com/v1",
"api_key": "sk-***",
@@ -78,7 +78,7 @@ GET /api/v1/asr
"id": "def67890",
"user_id": 1,
"name": "SenseVoice 中文识别",
"vendor": "SiliconFlow",
"vendor": "OpenAI Compatible",
"language": "zh",
"base_url": "https://api.siliconflow.cn/v1",
"api_key": "sf-***",
@@ -114,7 +114,7 @@ GET /api/v1/asr/{id}
"id": "abc12345",
"user_id": 1,
"name": "Whisper 多语种识别",
"vendor": "OpenAI",
"vendor": "OpenAI Compatible",
"language": "Multi-lingual",
"base_url": "https://api.openai.com/v1",
"api_key": "sk-***",
@@ -140,7 +140,7 @@ POST /api/v1/asr
```json
{
"name": "SenseVoice 中文识别",
"vendor": "SiliconFlow",
"vendor": "OpenAI Compatible",
"language": "zh",
"base_url": "https://api.siliconflow.cn/v1",
"api_key": "sk-your-api-key",
@@ -157,7 +157,7 @@ POST /api/v1/asr
| 字段 | 类型 | 必填 | 说明 |
|------|------|------|------|
| name | string | 是 | 模型显示名称 |
| vendor | string | 是 | 供应商: "OpenAI" / "SiliconFlow" / "Paraformer" |
| vendor | string | 是 | 供应商: "OpenAI Compatible" / "Paraformer" |
| language | string | 是 | 语言: "zh" / "en" / "Multi-lingual" |
| base_url | string | 是 | API Base URL |
| api_key | string | 是 | API Key |
@@ -347,7 +347,7 @@ class ASRTestResponse(BaseModel):
```json
{
"vendor": "OpenAI",
"vendor": "OpenAI Compatible",
"base_url": "https://api.openai.com/v1",
"api_key": "sk-xxx",
"model_name": "whisper-1",
@@ -357,11 +357,11 @@ class ASRTestResponse(BaseModel):
}
```
### SiliconFlow Paraformer
### OpenAI Compatible Paraformer
```json
{
"vendor": "SiliconFlow",
"vendor": "OpenAI Compatible",
"base_url": "https://api.siliconflow.cn/v1",
"api_key": "sf-xxx",
"model_name": "paraformer-v2",
@@ -393,7 +393,7 @@ class ASRTestResponse(BaseModel):
| test_filter_asr_models_by_language | 按语言过滤测试 |
| test_filter_asr_models_by_enabled | 按启用状态过滤测试 |
| test_create_asr_model_with_hotwords | 热词配置测试 |
| test_test_asr_model_siliconflow | SiliconFlow 供应商测试 |
| test_test_asr_model_siliconflow | OpenAI Compatible 供应商测试 |
| test_test_asr_model_openai | OpenAI 供应商测试 |
| test_different_asr_languages | 多语言测试 |
| test_different_asr_vendors | 多供应商测试 |

20
api/docs/llm.md
View File

@@ -20,7 +20,7 @@ interface LLMModel {
id: string; // 模型唯一标识 (8位UUID)
user_id: number; // 所属用户ID
name: string; // 模型显示名称
vendor: string; // 供应商: "OpenAI" | "SiliconFlow" | "Dify" | "FastGPT" | 等
vendor: string; // 供应商: "OpenAI Compatible" | "Dify" | "FastGPT" | 等
type: string; // 类型: "text" | "embedding" | "rerank"
base_url: string; // API Base URL
api_key: string; // API Key
@@ -64,7 +64,7 @@ GET /api/v1/llm
"id": "abc12345",
"user_id": 1,
"name": "GPT-4o",
"vendor": "OpenAI",
"vendor": "OpenAI Compatible",
"type": "text",
"base_url": "https://api.openai.com/v1",
"api_key": "sk-***",
@@ -79,7 +79,7 @@ GET /api/v1/llm
"id": "def67890",
"user_id": 1,
"name": "Embedding-3-Small",
"vendor": "OpenAI",
"vendor": "OpenAI Compatible",
"type": "embedding",
"base_url": "https://api.openai.com/v1",
"api_key": "sk-***",
@@ -111,7 +111,7 @@ GET /api/v1/llm/{id}
"id": "abc12345",
"user_id": 1,
"name": "GPT-4o",
"vendor": "OpenAI",
"vendor": "OpenAI Compatible",
"type": "text",
"base_url": "https://api.openai.com/v1",
"api_key": "sk-***",
@@ -137,7 +137,7 @@ POST /api/v1/llm
```json
{
"name": "GPT-4o",
"vendor": "OpenAI",
"vendor": "OpenAI Compatible",
"type": "text",
"base_url": "https://api.openai.com/v1",
"api_key": "sk-your-api-key",
@@ -314,11 +314,11 @@ class LLMModelTestResponse(BaseModel):
## 供应商配置示例
### OpenAI
### OpenAI Compatible (OpenAI Endpoint)
```json
{
"vendor": "OpenAI",
"vendor": "OpenAI Compatible",
"base_url": "https://api.openai.com/v1",
"api_key": "sk-xxx",
"model_name": "gpt-4o",
@@ -327,11 +327,11 @@ class LLMModelTestResponse(BaseModel):
}
```
### SiliconFlow
### OpenAI Compatible
```json
{
"vendor": "SiliconFlow",
"vendor": "OpenAI Compatible",
"base_url": "https://api.siliconflow.com/v1",
"api_key": "sf-xxx",
"model_name": "deepseek-v3",
@@ -356,7 +356,7 @@ class LLMModelTestResponse(BaseModel):
```json
{
"vendor": "OpenAI",
"vendor": "OpenAI Compatible",
"base_url": "https://api.openai.com/v1",
"api_key": "sk-xxx",
"model_name": "text-embedding-3-small",

9
api/docs/model-access.md
View File

@@ -20,7 +20,7 @@ interface LLMModel {
id: string; // 模型唯一标识
user_id: number; // 所属用户ID
name: string; // 模型显示名称
vendor: string; // 供应商: "OpenAI Compatible" | "SiliconFlow" | "Dify" | "FastGPT"
vendor: string; // 供应商: "OpenAI Compatible" | "Dify" | "FastGPT"
type: string; // 类型: "text" | "embedding" | "rerank"
base_url: string; // API Base URL
api_key: string; // API Key
@@ -57,7 +57,7 @@ interface TTSModel {
id: string;
user_id: number;
name: string;
vendor: string; // "Ali" | "Volcano" | "Minimax" | "硅基流动"
vendor: string; // "OpenAI Compatible" | "Ali" | "Volcano" | "Minimax"
language: string; // "zh" | "en"
voice_list?: string[]; // 支持的声音列表
enabled: boolean;
@@ -316,7 +316,6 @@ class LLMModelType(str, Enum):
class LLMModelVendor(str, Enum):
OPENAI_COMPATIBLE = "OpenAI Compatible"
SILICONFLOW = "SiliconFlow"
DIFY = "Dify"
FASTGPT = "FastGPT"
@@ -389,11 +388,11 @@ class ASRModelOut(ASRModelBase):
}
```
### SiliconFlow
### OpenAI Compatible
```json
{
"vendor": "SiliconFlow",
"vendor": "OpenAI Compatible",
"base_url": "https://api.siliconflow.com/v1",
"api_key": "sf-xxx",
"model_name": "deepseek-v3"

24
api/init_db.py
View File

@@ -135,21 +135,21 @@ def rebuild_vector_store(reset_doc_status: bool = True):
def init_default_data():
with db_session() as db:
# 检查是否已有数据
# SiliconFlow CosyVoice 2.0 预设声音 (8个)
# OpenAI Compatible (SiliconFlow API) CosyVoice 2.0 预设声音 (8个)
# 参考: https://docs.siliconflow.cn/cn/api-reference/audio/create-speech
voices = [
# 男声 (Male Voices)
Voice(id="alex", name="Alex", vendor="SiliconFlow", gender="Male", language="en",
Voice(id="alex", name="Alex", vendor="OpenAI Compatible", gender="Male", language="en",
description="Steady male voice.", is_system=True),
Voice(id="david", name="David", vendor="SiliconFlow", gender="Male", language="en",
Voice(id="david", name="David", vendor="OpenAI Compatible", gender="Male", language="en",
description="Cheerful male voice.", is_system=True),
# 女声 (Female Voices)
Voice(id="bella", name="Bella", vendor="SiliconFlow", gender="Female", language="en",
Voice(id="bella", name="Bella", vendor="OpenAI Compatible", gender="Female", language="en",
description="Passionate female voice.", is_system=True),
Voice(id="claire", name="Claire", vendor="SiliconFlow", gender="Female", language="en",
Voice(id="claire", name="Claire", vendor="OpenAI Compatible", gender="Female", language="en",
description="Gentle female voice.", is_system=True),
]
seed_if_empty(db, Voice, voices, "✅ 默认声音数据已初始化 (SiliconFlow CosyVoice 2.0)")
seed_if_empty(db, Voice, voices, "✅ 默认声音数据已初始化 (OpenAI Compatible CosyVoice 2.0)")
def init_default_tools(recreate: bool = False):
@@ -181,7 +181,7 @@ def init_default_assistants():
voice="anna",
speed=1.0,
hotwords=[],
tools=["calculator", "current_time"],
tools=["current_time"],
interruption_sensitivity=500,
config_mode="platform",
llm_model_id="deepseek-chat",
@@ -215,7 +215,7 @@ def init_default_assistants():
voice="alex",
speed=1.0,
hotwords=["grammar", "vocabulary", "practice"],
tools=["calculator"],
tools=["current_time"],
interruption_sensitivity=400,
config_mode="platform",
),
@@ -294,7 +294,7 @@ def init_default_llm_models():
id="deepseek-chat",
user_id=1,
name="DeepSeek Chat",
vendor="SiliconFlow",
vendor="OpenAI Compatible",
type="text",
base_url="https://api.deepseek.com",
api_key="YOUR_API_KEY", # 用户需替换
@@ -320,7 +320,7 @@ def init_default_llm_models():
id="text-embedding-3-small",
user_id=1,
name="Embedding 3 Small",
vendor="OpenAI",
vendor="OpenAI Compatible",
type="embedding",
base_url="https://api.openai.com/v1",
api_key="YOUR_API_KEY",
@@ -339,7 +339,7 @@ def init_default_asr_models():
id="FunAudioLLM/SenseVoiceSmall",
user_id=1,
name="FunAudioLLM/SenseVoiceSmall",
vendor="SiliconFlow",
vendor="OpenAI Compatible",
language="Multi-lingual",
base_url="https://api.siliconflow.cn/v1",
api_key="YOUR_API_KEY",
@@ -353,7 +353,7 @@ def init_default_asr_models():
id="TeleAI/TeleSpeechASR",
user_id=1,
name="TeleAI/TeleSpeechASR",
vendor="SiliconFlow",
vendor="OpenAI Compatible",
language="Multi-lingual",
base_url="https://api.siliconflow.cn/v1",
api_key="YOUR_API_KEY",