当前位置：首页 > article >正文

全自动PPT转视频解决方案：PPT + TTS + 自动化处理

article 2024/11/26 11:27:02

一、概述

本文介绍如何使用Python实现PPT自动转换为视频的完整流程，包括PPT处理、文本提取、语音合成和视频生成，全程无需人工干预。

二、所需环境和库

pip install python-pptx
pip install azure-cognitiveservices-speech
pip install moviepy
pip install pillow

三、完整代码实现

1. PPT文本提取

from pptx import Presentation

def extract_text_from_ppt(ppt_path):
    prs = Presentation(ppt_path)
    slides_text = []
    
    for slide in prs.slides:
        text_parts = []
        for shape in slide.shapes:
            if hasattr(shape, "text"):
                text_parts.append(shape.text)
        slides_text.append(" ".join(text_parts))
    
    return slides_text

2. 语音合成模块

from azure.cognitiveservices.speech import *
import os

class TTSGenerator:
    def __init__(self, subscription_key, region):
        self.speech_config = SpeechConfig(
            subscription=subscription_key, 
            region=region
        )
        self.speech_config.speech_synthesis_voice_name = "zh-CN-XiaoxiaoNeural"
    
    def generate_audio(self, text, output_path):
        audio_config = AudioConfig(filename=output_path)
        synthesizer = SpeechSynthesizer(
            speech_config=self.speech_config, 
            audio_config=audio_config
        )
        synthesizer.speak_text_async(text).get()

3. PPT转图片

import win32com.client
import os

def convert_ppt_to_images(ppt_path, output_dir):
    powerpoint = win32com.client.Dispatch("Powerpoint.Application")
    presentation = powerpoint.Presentations.Open(ppt_path)
    
    # 确保输出目录存在
    os.makedirs(output_dir, exist_ok=True)
    
    try:
        presentation.SaveAs(
            os.path.join(output_dir, "slide"),
            17  # ppSaveAsJPG
        )
    finally:
        presentation.Close()
        powerpoint.Quit()

4. 视频生成

from moviepy.editor import *
import glob

def create_video(image_dir, audio_files, output_path):
    # 获取所有图片和音频文件
    images = sorted(glob.glob(os.path.join(image_dir, "*.jpg")))
    
    # 创建视频片段
    clips = []
    for img, audio in zip(images, audio_files):
        # 获取音频时长
        audio_clip = AudioFileClip(audio)
        duration = audio_clip.duration
        
        # 创建图片视频片段
        video_clip = ImageClip(img).set_duration(duration)
        video_clip = video_clip.set_audio(audio_clip)
        clips.append(video_clip)
    
    # 合并所有片段
    final_clip = concatenate_videoclips(clips)
    
    # 导出视频
    final_clip.write_videofile(
        output_path,
        fps=24,
        codec='libx264',
        audio_codec='aac'
    )

5. 主程序

def main():
    # 配置参数
    ppt_path = "presentation.pptx"
    output_dir = "output"
    azure_key = "你的Azure密钥"
    azure_region = "你的区域"
    
    # 创建输出目录
    os.makedirs(output_dir, exist_ok=True)
    
    # 1. 提取PPT文本
    slides_text = extract_text_from_ppt(ppt_path)
    
    # 2. 初始化TTS生成器
    tts = TTSGenerator(azure_key, azure_region)
    
    # 3. 生成音频文件
    audio_files = []
    for i, text in enumerate(slides_text):
        audio_path = os.path.join(output_dir, f"audio_{i}.wav")
        tts.generate_audio(text, audio_path)
        audio_files.append(audio_path)
    
    # 4. 转换PPT为图片
    image_dir = os.path.join(output_dir, "slides")
    convert_ppt_to_images(ppt_path, image_dir)
    
    # 5. 生成最终视频
    create_video(image_dir, audio_files, "final_video.mp4")

if __name__ == "__main__":
    main()