更新代码，添加ocr和whisper

code/simple_ocr_test.py

@@ -1,137 +0,0 @@
-#!/usr/bin/env python3
-# -*- coding: utf-8 -*-
-"""
-简化OCR测试脚本
-"""
-
-import cv2
-import json
-from pathlib import Path
-
-def test_paddleocr():
-    """测试PaddleOCR基本功能"""
-    print("🔧 测试PaddleOCR...")
-    
-    try:
-        from paddleocr import PaddleOCR
-        
-        # 初始化OCR
-        ocr = PaddleOCR(
-            use_textline_orientation=True,
-            lang='ch'
-        )
-        print("✅ PaddleOCR初始化成功")
-        
-        # 测试视频文件
-        video_path = "/root/autodl-tmp/hot_video_analyse/video2audio/sample_demo_6.mp4"
-        
-        # 读取一帧进行测试
-        cap = cv2.VideoCapture(video_path)
-        ret, frame = cap.read()
-        cap.release()
-        
-        if not ret:
-            print("❌ 无法读取视频帧")
-            return False
-        
-        print("📸 成功读取视频帧")
-        
-        # 进行OCR识别 - 使用predict方法
-        try:
-            result = ocr.predict(frame)
-        except:
-            # 如果predict不工作，尝试ocr方法
-            result = ocr.ocr(frame)
-        
-        print(f"🔍 OCR结果类型: {type(result)}")
-        
-        # 处理结果
-        if result:
-            if isinstance(result, list) and len(result) > 0:
-                detections = result[0] if isinstance(result[0], list) else result
-                print(f"🔍 OCR结果: {len(detections)} 个检测")
-                
-                for i, detection in enumerate(detections[:3]):  # 只显示前3个
-                    if detection and len(detection) >= 2:
-                        try:
-                            bbox, text_info = detection
-                            if isinstance(text_info, tuple) and len(text_info) >= 2:
-                                text, confidence = text_info
-                                print(f"  {i+1}. 文本: '{text}' 置信度: {confidence:.3f}")
-                            else:
-                                print(f"  {i+1}. 检测到文本但格式异常: {text_info}")
-                        except Exception as e:
-                            print(f"  {i+1}. 处理检测结果时出错: {e}")
-            else:
-                print("🔍 OCR结果为空或格式异常")
-        else:
-            print("🔍 OCR结果为None")
-        
-        return True
-        
-    except Exception as e:
-        print(f"❌ PaddleOCR测试失败: {e}")
-        import traceback
-        print(f"详细错误: {traceback.format_exc()}")
-        return False
-
-def test_easyocr():
-    """测试EasyOCR基本功能"""
-    print("\n🔧 测试EasyOCR...")
-    
-    try:
-        import easyocr
-        
-        # 初始化OCR
-        reader = easyocr.Reader(['ch_sim', 'en'])
-        print("✅ EasyOCR初始化成功")
-        
-        # 测试视频文件
-        video_path = "/root/autodl-tmp/hot_video_analyse/video2audio/sample_demo_6.mp4"
-        
-        # 读取一帧进行测试
-        cap = cv2.VideoCapture(video_path)
-        ret, frame = cap.read()
-        cap.release()
-        
-        if not ret:
-            print("❌ 无法读取视频帧")
-            return
-        
-        print("📸 成功读取视频帧")
-        
-        # 进行OCR识别
-        result = reader.readtext(frame)
-        
-        print(f"🔍 OCR结果: {len(result)} 个检测")
-        
-        for i, detection in enumerate(result[:3]):  # 只显示前3个
-            bbox, text, confidence = detection
-            print(f"  {i+1}. 文本: '{text}' 置信度: {confidence:.3f}")
-        
-        return True
-        
-    except Exception as e:
-        print(f"❌ EasyOCR测试失败: {e}")
-        return False
-
-def main():
-    """主函数"""
-    print("🧪 简化OCR功能测试")
-    print("=" * 40)
-    
-    paddle_ok = test_paddleocr()
-    easy_ok = test_easyocr()
-    
-    print("\n" + "=" * 40)
-    print("📊 测试结果:")
-    print(f"  PaddleOCR: {'✅ 成功' if paddle_ok else '❌ 失败'}")
-    print(f"  EasyOCR:   {'✅ 成功' if easy_ok else '❌ 失败'}")
-    
-    if paddle_ok or easy_ok:
-        print("\n🎉 至少有一个OCR引擎工作正常，可以继续使用！")
-    else:
-        print("\n⚠️ 所有OCR引擎都有问题，请检查安装")
-
-if __name__ == "__main__":
-    main()