百度MP3+图片+文字：生成结果文件；（声音58秒，视频59秒，同步性需要进一步优化）

import os

os_sep = os.sep
this_file_abspath = os.path.abspath(__file__)
this_file_dirname, this_file_name = os.path.dirname(this_file_abspath), os.path.abspath(__file__).split(os_sep)[
    -1]

f_mp3 = '{}{}{}'.format(this_file_dirname, os_sep, 'auido.mp3')
from playsound import playsound

import time
import math

this_time = time.time()

# playsound(f_mp3)

# t_spend = time.time() - this_time
t_spend = 58.777058839797974
# 音频的秒数
t_spend = math.ceil(t_spend)
import cv2
import glob

'''
python+opencv视频图像相互转换 - CSDN博客 https://blog.csdn.net/m0_37733057/article/details/79023693
链接：https://www.zhihu.com/question/49558804/answer/343058915

OpenCV: Drawing Functions in OpenCV https://docs.opencv.org/3.1.0/dc/da5/tutorial_py_drawing_functions.html

'''
# 每秒传输帧数(Frames Per Second)
fps = 100  # 保存视频的FPS，可以适当调整 FPS是图像领域中的定义，是指画面每秒传输帧数，通俗来讲就是指动画或视频的画面数。FPS是测量用于保存、显示动态视频的信息数量。每秒钟帧数愈多，所显示的动作就会愈流畅。通常，要避免动作不流畅的最低是30。某些计算机视频格式，每秒只能提供15帧。
fps = 15
fps = 5

fourcc = cv2.VideoWriter_fourcc('M', 'J', 'P', 'G')  # opencv3.0

f_v = '{}{}'.format(int(time.time()), 'saveVideo.avi')
f_img_d = '{}{}{}{}{}'.format(this_file_dirname, os_sep, 'mypng', os_sep, '*.jpg')
imgs = glob.glob(f_img_d)

img = cv2.imread(imgs[0])
img_size = (img.shape[1], img.shape[0])
videoWriter = cv2.VideoWriter(f_v, fourcc, fps, img_size)

"""
用图片总数均分音频时间
"""
os_delay_factor = 0.14
os_delay_factor = 0.11
myinterval = t_spend / len(imgs) * os_delay_factor

f, l = 'mybaidu.parp.txt', []
with open(f, 'r', encoding='utf-8') as fr:
    for i in fr:
        ii = i.replace('\n', '')
        l.append(ii)
char_loop_l = []
for i in l:
    mystr, le = '', len(i)
    for ii in range(le):
        iii = i[ii]
        print('-----', iii)
        mystr = '{}{}'.format(mystr, iii)
        print(mystr)
        char_loop_l.append(iii)

#
# from fontTools.ttLib import TTFont
# myfont = TTFont('simhei.ttf')

from PIL import Image, ImageDraw, ImageFont

myfont = ImageFont.truetype("simhei.ttf", 50, encoding="utf-8")

import cv2
import numpy as np

os_delay_factor = 0.23
myinterval = t_spend / (len(char_loop_l) * 1) * os_delay_factor
for i in l:
    i_index=l.index(i)
    img_index=i_index%len(imgs)
    imgname=imgs[img_index]

    mystr, le = '', len(i)
    for ii in range(le):
        iii = i[ii]
        print('-----', iii)
        if len(mystr) % 15 == 0:
            mystr = '{}{}'.format(mystr, '\n')
        mystr = '{}{}'.format(mystr, iii)
        print(mystr)

        this_time = time.time()
        while time.time() - this_time < myinterval:
            print(imgname)

            frame = cv2.imread(imgname)
            frame_cv2 = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
            frame_pil = Image.fromarray(frame_cv2)  # 转为PIL的图片格式

            draw = ImageDraw.Draw(frame_pil)
            font = ImageFont.truetype("simhei.ttf", 50, encoding="utf-8")
            # 第一个参数为字体，中文黑体
            # 第二个为字体大小
            ImageDraw.Draw(frame_pil).text((100, 20), mystr, (0, 0, 255), font)

            '''
                frame_pil:目标图像
                第一个参数为打印的坐标
                第二个为打印的文本
                第三个为字体颜色
                第四个为字体
            '''

            frame_cv2 = cv2.cvtColor(np.array(frame_pil), cv2.COLOR_RGB2BGR)
            img = frame_cv2
            videoWriter.write(img)

videoWriter.release()

　　

原理是将字幕写到图片中，然后再写入视频的帧中

对图片适应

import os, time, glob
import cv2

os_sep = os.sep
this_file_abspath = os.path.abspath(__file__)
this_file_dirname, this_file_name = os.path.dirname(this_file_abspath), os.path.abspath(__file__).split(os_sep)[
    -1]

f_v = '{}{}'.format(int(time.time()), 'saveVideo.avi')
f_img_d = '{}{}{}{}{}'.format(this_file_dirname, os_sep, 'mypng', os_sep, '*.jpg')
imgs, img_size_d = glob.glob(f_img_d), {}
for i in imgs:
    img = cv2.imread(i)
    w_h_s = '{},{}'.format(img.shape[1], img.shape[0])
    if w_h_s not in img_size_d:
        img_size_d[w_h_s] = 1
    else:
        img_size_d[w_h_s] += 1

mode_img_size_wh = [int(i) for i in sorted(img_size_d.items(), key=lambda img_size_d: img_size_d[0], reverse=True)[0][0].split(',')]

import os

os_sep = os.sep
this_file_abspath = os.path.abspath(__file__)
this_file_dirname, this_file_name = os.path.dirname(this_file_abspath), os.path.abspath(__file__).split(os_sep)[-1]

f_mp3 = '{}{}{}'.format(this_file_dirname, os_sep, 'auido.mp3')
from playsound import playsound

import time
import math

this_time = time.time()

# playsound(f_mp3)

# t_spend = time.time() - this_time
t_spend = 58.777058839797974
# 音频的秒数
t_spend = math.ceil(t_spend)
import cv2
import glob

'''
python+opencv视频图像相互转换 - CSDN博客 https://blog.csdn.net/m0_37733057/article/details/79023693
链接：https://www.zhihu.com/question/49558804/answer/343058915

OpenCV: Drawing Functions in OpenCV https://docs.opencv.org/3.1.0/dc/da5/tutorial_py_drawing_functions.html

'''
# 每秒传输帧数(Frames Per Second)
fps = 100  # 保存视频的FPS，可以适当调整 FPS是图像领域中的定义，是指画面每秒传输帧数，通俗来讲就是指动画或视频的画面数。FPS是测量用于保存、显示动态视频的信息数量。每秒钟帧数愈多，所显示的动作就会愈流畅。通常，要避免动作不流畅的最低是30。某些计算机视频格式，每秒只能提供15帧。

fps, fourcc = 15, cv2.VideoWriter_fourcc('M', 'J', 'P', 'G')
# fourcc = cv2.VideoWriter_fourcc('M', 'J', 'P', 'G')  # opencv3.0
videoWriter = cv2.VideoWriter(f_v, fourcc, fps, (mode_img_size_wh[0], mode_img_size_wh[1]))

f_v = '{}{}'.format(int(time.time()), 'saveVideo.avi')
f_img_d = '{}{}{}{}{}'.format(this_file_dirname, os_sep, 'mypng', os_sep, '*.jpg')
imgs = glob.glob(f_img_d)

"""
用图片总数均分音频时间
"""
os_delay_factor = 0.14
os_delay_factor = 0.11
myinterval = t_spend / len(imgs) * os_delay_factor

f, l = 'mybaidu.parp.txt', []
with open(f, 'r', encoding='utf-8') as fr:
    for i in fr:
        ii = i.replace('\n', '')
        l.append(ii)
char_loop_l = []
for i in l:
    mystr, le = '', len(i)
    for ii in range(le):
        iii = i[ii]
        print('-----', iii)
        mystr = '{}{}'.format(mystr, iii)
        print(mystr)
        char_loop_l.append(iii)

#
# from fontTools.ttLib import TTFont
# myfont = TTFont('simhei.ttf')

def resize_rescale_pilimg(img_f, w_h_tuple=(mode_img_size_wh[0],mode_img_size_wh[1]), mid_factor=1):
    img_n,img_type=img_f.split('.')[-2],img_f.split('.')[-1]
    img_type = 'PNG'
    img_f_new = '{}{}{}'.format(img_n, 'resize_rescale.', img_type)
    mid_icon = Image.open(img_f)
    mid_icon_w, mid_icon_h = w_h_tuple[0] * mid_factor, w_h_tuple[1] * mid_factor
    mid_icon = mid_icon.resize((mid_icon_w, mid_icon_h), Image.ANTIALIAS)
    mid_icon.save(img_f_new,img_type)
    return img_f_new

from PIL import Image, ImageDraw, ImageFont

myfont = ImageFont.truetype("simhei.ttf", 50, encoding="utf-8")

import cv2
import numpy as np

os_delay_factor = 0.245
myinterval = t_spend / (len(char_loop_l) * 1) * os_delay_factor
for i in l:
    i_index = l.index(i)
    img_index = i_index % len(imgs)
    imgname = imgs[img_index]

    mystr, le = '', len(i)
    for ii in range(le):
        iii = i[ii]
        print('-----', iii)
        if len(mystr) % 15 == 0:
            mystr = '{}{}'.format(mystr, '\n')
        mystr = '{}{}'.format(mystr, iii)
        print(mystr)

        this_time = time.time()
        while time.time() - this_time < myinterval:
            print(imgname)

            frame = cv2.imread(imgname)

            if (frame.shape[1], frame.shape[0]) != (mode_img_size_wh[0], mode_img_size_wh[1]):
                imgname = resize_rescale_pilimg(imgname)
                frame = cv2.imread(imgname)
            else:
                pass

            frame_cv2 = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
            frame_pil = Image.fromarray(frame_cv2)  # 转为PIL的图片格式

            draw = ImageDraw.Draw(frame_pil)
            font = ImageFont.truetype("simhei.ttf", 50, encoding="utf-8")
            # 第一个参数为字体，中文黑体
            # 第二个为字体大小
            ImageDraw.Draw(frame_pil).text((100, 20), mystr, (0, 0, 255), font)

            '''
                frame_pil:目标图像
                第一个参数为打印的坐标
                第二个为打印的文本
                第三个为字体颜色
                第四个为字体
            '''

            frame_cv2 = cv2.cvtColor(np.array(frame_pil), cv2.COLOR_RGB2BGR)
            img = frame_cv2
            videoWriter.write(img)

videoWriter.release()

'''
待解决：
0-同一批的图片的尺寸不同
1-视频、音频播放时长的同步系数的自计算
2-

字幕

'''

from PIL import Image

#
#
# logo = 'fugu.png'
# mid_icon = Image.open(logo)
# mid_icon_w, mid_icon_h = mid_icon.size
# mid_factor = 5
# mid_icon_w, mid_icon_h = mid_icon_w * mid_factor, mid_icon_h * mid_factor
# mid_icon = mid_icon.resize((mid_icon_w * 3, mid_icon_h), Image.ANTIALIAS)
# mid_icon.save('fugu.2.png', 'PNG')