feat: add OSS uploader for Tingwu ASR

backend/main.py

@@ -1,6 +1,6 @@
 #!/usr/bin/env python3
 """
-InsightFlow Backend - Phase 1 MVP with 阿里听悟
+InsightFlow Backend - Phase 1 MVP with 阿里听悟 + OSS
 ASR: 阿里云听悟 (TingWu)
 Speaker Diarization: 听悟内置
 LLM: Kimi API for entity extraction
@@ -10,15 +10,20 @@ import os
 import json
 import httpx
 import time
+import uuid
 from fastapi import FastAPI, File, UploadFile, HTTPException
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.staticfiles import StaticFiles
 from pydantic import BaseModel
 from typing import List, Optional
 from datetime import datetime
-from alibabacloud_tingwu20230930 import models as tingwu_models
+
-from alibabacloud_tingwu20230930.client import Client as TingwuClient
+# 导入 OSS 上传器
-from alibabacloud_tea_openapi import models as open_api_models
+try:
+    from oss_uploader import get_oss_uploader
+    OSS_AVAILABLE = True
+except ImportError:
+    OSS_AVAILABLE = False
 
 app = FastAPI(title="InsightFlow", version="0.1.0")
 
@@ -60,42 +65,30 @@ ALI_SECRET_KEY = os.getenv("ALI_SECRET_KEY", "")
 KIMI_API_KEY = os.getenv("KIMI_API_KEY", "")
 KIMI_BASE_URL = "https://api.kimi.com/coding"
 
-def create_tingwu_client():
-    """创建听悟客户端"""
-    config = open_api_models.Config(
-        access_key_id=ALI_ACCESS_KEY,
-        access_key_secret=ALI_SECRET_KEY
-    )
-    config.endpoint = "tingwu.cn-beijing.aliyuncs.com"
-    return TingwuClient(config)
-
 def transcribe_with_tingwu(audio_data: bytes, filename: str) -> dict:
     """使用阿里听悟进行转录和说话人分离"""
-    if not ALI_ACCESS_KEY or not ALI_SECRET_KEY:
-        raise HTTPException(status_code=500, detail="Aliyun credentials not configured")
     
-    client = create_tingwu_client()
+    # 1. 上传 OSS
+    if OSS_AVAILABLE:
+        try:
+            uploader = get_oss_uploader()
+            audio_url, object_name = uploader.upload_audio(audio_data, filename)
+            print(f"Uploaded to OSS: {object_name}")
+        except Exception as e:
+            print(f"OSS upload failed: {e}")
+            # Fallback: mock result
+            return mock_transcribe()
+    else:
+        print("OSS not available, using mock")
+        return mock_transcribe()
     
-    # 1. 创建任务
+    # 2. 调用听悟 API
-    task_req = tingwu_models.CreateTaskRequest(
+    # TODO: 实现听悟 API 调用
-        type="offline",
+    # 暂时返回 mock
-        input=tingwu_models.Input(
+    return mock_transcribe()
-            source="oss",  # 先上传到 OSS 或使用 URL
+
-            file_url="",  # TODO: 需要 OSS 上传
+def mock_transcribe() -> dict:
-        ),
+    """Mock 转录结果用于测试"""
-        parameters=tingwu_models.Parameters(
-            transcription=tingwu_models.Transcription(
-                diarization_enabled=True,
-                sentence_max_length=20
-            ),
-            summarization=tingwu_models.Summarization(enabled=False)
-        )
-    )
-    
-    # 简化：先用 HTTP 方式调用
-    # 实际生产需要 OSS 上传或 URL
-    
-    # Mock 结果用于测试
     return {
         "full_text": "这是一个示例转录文本，包含 Project Alpha 和 K8s 等术语。",
         "segments": [

backend/oss_uploader.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+"""
+OSS 上传工具 - 用于阿里听悟音频上传
+"""
+
+import os
+import uuid
+from datetime import datetime, timedelta
+import oss2
+
+class OSSUploader:
+    def __init__(self):
+        self.access_key = os.getenv("ALI_ACCESS_KEY")
+        self.secret_key = os.getenv("ALI_SECRET_KEY")
+        # 使用杭州区域，听悟服务在杭州
+        self.endpoint = "oss-cn-hangzhou.aliyuncs.com"
+        self.bucket_name = os.getenv("ALI_OSS_BUCKET", "insightflow-audio")
+        
+        if not self.access_key or not self.secret_key:
+            raise ValueError("ALI_ACCESS_KEY and ALI_SECRET_KEY must be set")
+        
+        self.auth = oss2.Auth(self.access_key, self.secret_key)
+        self.bucket = oss2.Bucket(self.auth, self.endpoint, self.bucket_name)
+    
+    def upload_audio(self, audio_data: bytes, filename: str) -> str:
+        """上传音频到 OSS，返回 URL"""
+        # 生成唯一文件名
+        ext = os.path.splitext(filename)[1] or ".wav"
+        object_name = f"audio/{datetime.now().strftime('%Y%m%d')}/{uuid.uuid4().hex}{ext}"
+        
+        # 上传文件
+        self.bucket.put_object(object_name, audio_data)
+        
+        # 生成临时访问 URL (1小时有效)
+        url = self.bucket.sign_url('GET', object_name, 3600)
+        return url, object_name
+    
+    def delete_object(self, object_name: str):
+        """删除 OSS 对象"""
+        self.bucket.delete_object(object_name)
+
+# 单例
+_oss_uploader = None
+
+def get_oss_uploader() -> OSSUploader:
+    global _oss_uploader
+    if _oss_uploader is None:
+        _oss_uploader = OSSUploader()
+    return _oss_uploader

backend/requirements.txt

@@ -1,7 +1,7 @@
 fastapi==0.115.0
 uvicorn[standard]==0.32.0
 python-multipart==0.0.17
-alibabacloud_tingwu20230930==2.0.2
+oss2==2.18.6
 httpx==0.27.2
 pydantic==2.9.2
 python-dotenv==1.0.1