pip install moviepy pydub

from moviepy.editor import VideoFileClip
from pydub import AudioSegment
from pydub.silence import split_on_silence
import os

def extract_and_cut_audio(video_path, output_mp3="output.mp3"):
    """
    提取MP4中的音頻，切割掉靜音部分，保存為MP3
    
    參數(shù):
        video_path: MP4視頻文件路徑
        output_mp3: 輸出的MP3文件路徑
    """
    # 1. 從MP4中提取音頻并保存為臨時WAV（pydub處理WAV更方便）
    temp_wav = "temp_audio.wav"
    with VideoFileClip(video_path) as video:
        audio = video.audio
        audio.write_audiofile(temp_wav, codec="pcm_s16le")  # 保存為WAV
    
    # 2. 加載音頻并分析靜音部分
    audio = AudioSegment.from_wav(temp_wav)
    
    # 配置靜音檢測參數(shù)（可根據(jù)實際情況調整）
    silence_thresh = -50  # 靜音閾值（dBFS），值越小越敏感
    min_silence_len = 1000  # 最小靜音時長（毫秒）
    keep_silence = 200  # 保留靜音的邊緣（毫秒），避免切割太生硬
    
    # 按靜音分割音頻
    non_silent_parts = split_on_silence(
        audio,
        min_silence_len=min_silence_len,
        silence_thresh=silence_thresh,
        keep_silence=keep_silence
    )
    
    if not non_silent_parts:
        print("未檢測到有聲音的片段")
        return
    
    # 3. 合并所有有聲音的片段
    combined = AudioSegment.empty()
    for part in non_silent_parts:
        combined += part
    
    # 4. 保存為MP3
    combined.export(output_mp3, format="mp3")
    print(f"處理完成，輸出文件：{output_mp3}")
    
    # 清理臨時文件
    os.remove(temp_wav)

extract_and_cut_audio("meeting.mp4", "processed_meeting.mp3")