pip install opencv-python moviepy Pillow

import cv2
import numpy as np
from moviepy.editor import AudioFileClip, VideoFileClip
from pydub import AudioSegment
from PIL import Image, ImageDraw, ImageFont
import tempfile
import os
import re 

def replace_punctuation_with_at(input_string):
    # 使用正則表達(dá)式匹配所有標(biāo)點(diǎn)符號(hào)并替換為 @
    result = re.sub(r'[^\w\s]', '@', input_string)
    return result

def add_chinese_subtitle_to_frame(frame, subtitle_text, position, font_path, font_size, font_color):
    # 將OpenCV圖像轉(zhuǎn)換為PIL圖像
    img_pil = Image.fromarray(cv2.cvtColor(frame, cv2.COLOR_BGR2RGB))
    draw = ImageDraw.Draw(img_pil)
    font = ImageFont.truetype(font_path, font_size)
    
    # 獲取字幕文本的寬度和高度
    text_bbox = draw.textbbox((0, 0), subtitle_text, font=font)
    text_width = text_bbox[2] - text_bbox[0]
    text_height = text_bbox[3] - text_bbox[1]
    
    # 計(jì)算字幕的放置位置
    x = position[0] - text_width // 2
    y = position[1] - text_height // 2
    
    # 在PIL圖像上添加字幕
    draw.text((x, y), subtitle_text, font=font, fill=font_color)
    
    # 將PIL圖像轉(zhuǎn)換回OpenCV圖像
    frame = cv2.cvtColor(np.array(img_pil), cv2.COLOR_RGB2BGR)
    return frame

def sync_audio_video_add_subtitle(audio_path, video_path, output_path, subtitle_text, font_path, font_size=24, font_color=(255, 255, 255), subtitle_bottom_margin=30, audio_volume=1.0):
    # 加載音頻文件并調(diào)整音量
    original_audio = AudioSegment.from_file(audio_path)
    original_audio = original_audio + (audio_volume * 10 - 10)  # 調(diào)整音量
    silence = AudioSegment.silent(duration=500)  # 0.5秒的靜音
    audio_with_silence = silence + original_audio + silence
    
    # 創(chuàng)建臨時(shí)文件以保存修改后的音頻
    temp_audio_path = os.path.join(tempfile.gettempdir(), "temp_audio.mp3")
    audio_with_silence.export(temp_audio_path, format="mp3")

    # 加載修改后的音頻文件
    audio_clip = AudioFileClip(temp_audio_path)
    audio_duration = audio_clip.duration

    # 加載視頻文件
    cap = cv2.VideoCapture(video_path)
    fps = cap.get(cv2.CAP_PROP_FPS)
    frame_count = int(cap.get(cv2.CAP_PROP_FRAME_COUNT))
    video_duration = frame_count / fps

    # 計(jì)算新的視頻幀率
    new_fps = fps * (video_duration / audio_duration)

    # 獲取視頻尺寸
    width = int(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
    height = int(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))

    # 創(chuàng)建臨時(shí)文件以存儲(chǔ)中間視頻結(jié)果
    temp_video_path = os.path.join(tempfile.gettempdir(), "temp_video.mp4")

    # 創(chuàng)建VideoWriter對(duì)象
    out = cv2.VideoWriter(temp_video_path, cv2.VideoWriter_fourcc(*'mp4v'), new_fps, (width, height))

    # 分批讀取和寫入視頻幀，并添加字幕
    subtitle_text = replace_punctuation_with_at(subtitle_text)
    subtitle_text_list = [text for text in subtitle_text.split("@") if text]
    print(subtitle_text_list)
    subtitles_per_frame = frame_count // len(subtitle_text_list)
    current_subtitle_index = 0

    for frame_idx in range(frame_count):
        ret, frame = cap.read()
        if not ret:
            break
        # 添加當(dāng)前字幕到幀
        if frame_idx // subtitles_per_frame >= current_subtitle_index and current_subtitle_index < len(subtitle_text_list):
            frame = add_chinese_subtitle_to_frame(frame, subtitle_text_list[current_subtitle_index], (width // 2, height - subtitle_bottom_margin), font_path, font_size, font_color)
            if frame_idx // subtitles_per_frame > current_subtitle_index:
                current_subtitle_index += 1
        out.write(frame)

    cap.release()
    out.release()

    # 使用MoviePy將音頻和調(diào)整后的視頻合并
    video_clip = VideoFileClip(temp_video_path).set_duration(audio_duration)
    final_clip = video_clip.set_audio(audio_clip)
    
    # Trim the last 0.3 seconds
    final_clip = final_clip.subclip(0, final_clip.duration - 0.3)
    
    # Write the final video file
    final_clip.write_videofile(output_path, codec="libx264", audio_codec="aac")
    
    # Close the clips to release the file
    final_clip.close()
    video_clip.close()
    audio_clip.close()

    # 刪除臨時(shí)文件
    os.remove(temp_audio_path)
    os.remove(temp_video_path)

if __name__ == "__main__":
    # 示例用法
    audio_path = r"C:\Users\60568\Pictures\create\屈原\mp3\00000002.mp3"
    video_path = r"C:\Users\60568\Pictures\create\屈原\mp4\03.mp4"
    subtitle_text='然而屈原的直言進(jìn)諫,引來(lái)了朝中權(quán)臣的嫉恨,他成為了政治斗爭(zhēng)的犧牲品。'
    output_path = "synced_video.mp4"

    sync_audio_video_add_subtitle(audio_path, 
                                video_path, 
                                output_path,
                                subtitle_text=subtitle_text,
                                font_path="./NotoSansCJKsc-Regular.ttf",
                                font_size=40, # 設(shè)置字體大小
                                font_color=(255, 255, 255), # 設(shè)置字體顏色
                                subtitle_bottom_margin=80, # 設(shè)置字幕底部的位置
                                audio_volume=2) # 調(diào)整音頻音量，1.0為原始音量，1.5為增加50%音量

new_fps = fps * (video_duration / audio_duration)