def extract_frames(video_path, output_dir, frame_rate=1):"""从视频中抽取帧并保存到指定目录
参数:video_path (str): 视频文件的路径output_dir (str): 保存抽取帧的目录frame_rate (int): 每秒抽取的帧数，默认为1帧每秒
返回:int: 抽取的帧数"""# 如果输出目录不存在，则创建if not os.path.exists(output_dir):os.makedirs(output_dir)
# 打开视频文件cap = cv2.VideoCapture(video_path)
# 获取视频的帧率video_fps = cap.get(cv2.CAP_PROP_FPS)
# 计算每隔多少帧抽取一次frame_interval = int(video_fps / frame_rate)
frame_count = 0saved_frame_count = 0
while cap.isOpened():ret, frame = cap.read()if not ret:break
# 如果当前帧是需要保存的帧if frame_count % frame_interval == 0:frame_filename = os.path.join(output_dir, f"frame_{saved_frame_count:04d}.jpg")cv2.imwrite(frame_filename, frame)saved_frame_count += 1
frame_count += 1
# 释放视频捕获对象cap.release()
return saved_frame_count

def get_result(spk_content,images):# 中文 Prompt 指令prompt = ("你好，图中"+spk_content)base64_images = [encode_image(image) for image in images]# 组织用户消息user_content = [{"type": "text", "text": prompt}]base64_images = [{"type": "image_url","image_url": {"url": f"data:image/jpeg;base64,{base64_image}","detail": "high",},}for base64_image in base64_images]user_content.extend(base64_images)messages_template = [{"role": "user", "content": user_content}]# 构造请求参数
# OpenAI API Keyresponse = openai.ChatCompletion.create(model="gpt-4o",messages=messages_template,n=1,max_tokens=1024,temperature=0,seed=42,)# print(response.json())result = response.choices[0].message.contentreturn result

async def get_voice(text):"""Main function"""VOICE = "zh-CN-XiaoxiaoNeural"output_file = "response"+str(time())+".mp3"communicate = edge_tts.Communicate(text, VOICE)await communicate.save(output_file)return output_file

def save_video(video):# 获取当前时间戳timestamp = datetime.datetime.now().strftime("%Y%m%d%H%M%S")# 生成保存文件的路径video_path = os.path.join(SAVE_DIR, f"video_{timestamp}.mp4")# 将视频保存到文件with open(video, "rb") as f:video_data = f.read()with open(video_path, "wb") as f:f.write(video_data)
file_name = str(time())output_dir = "pic/"+file_name# 替换为你希望保存帧的目录frame_rate = 3# 每秒抽取1帧
num_frames = extract_frames(video_path, output_dir, frame_rate)print(f"总共抽取了 {num_frames} 帧")
spk_content = get_spk_txt(video_path)print("我的语音：",spk_content)images = [output_dir+"/"+i for i in os.listdir(output_dir)]resp = get_result(spk_content,images)# 生成语音文件# audio_file = get_voice(resp)audio_file = asyncio.run(get_voice(resp))print("语音合成完成")return resp, os.path.join(os.getcwd(), audio_file) if audio_file else None