MAC使用python下载字幕

安装软件

brew install tesseract

安装pytesseract和pillow

pip install pytesseract
pip install pillow

pip install SpeechRecognition

pip3 install soundfile

pip3 install torch 

pip3 install whisper

脚本参考：

python 复制代码

import os
import sys

import cv2
import pytesseract
import speech_recognition as sr
from moviepy.video.io.VideoFileClip import VideoFileClip
from pydub import AudioSegment


def extract_subtitles(video_path, output_dir):
    audio = AudioSegment.from_file(video_path, format='mp4')
    audio.export("audio.wav", format="wav")

    r = sr.Recognizer()
    audio_file = sr.AudioFile('audio.wav')
    with audio_file as source:
        audio = r.record(source)
    text = r.recognize_whisper(audio)

    print(text)


def extract_subtitles_v2(video_path, output_dir):
    r = sr.Recognizer()
    clip = VideoFileClip(video_path)
    clip.audio.write_audiofile('audio.wav')
    sound = AudioSegment.from_file('audio.wav', format='wav')

    def transcribe_audio(sound):
        transcript = ''
        with sr.AudioFile(sound) as source:
            audio_text = r.record(source)
            try:
                transcript = r.recognize_whisper(audio_text)
            except sr.UnknownValueError as e:
                print(e)
        return transcript

    transcription = transcribe_audio(sound)
    print(transcription)


def extract_subtitles_v1(video_path, output_dir):
    vidcap = cv2.VideoCapture(video_path)

    frames = []
    success, image = vidcap.read()

    count = 0
    success = True

    while success:
        frames.append(image)
        success, image = vidcap.read()
        count += 1

    # lang='chi_sim'
    for frame in frames:
        # text = pytesseract.image_to_string(frame, lang='eng')
        text = pytesseract.image_to_string(frame, lang='chi_sim')
        print(text)


def extract_subtitles_v3(video_path, output_dir):
    vidcap = cv2.VideoCapture(video_path)

    frames = []
    success, image = vidcap.read()

    count = 0
    success = True

    while success:
        success, image = vidcap.read()
        text = pytesseract.image_to_string(image, lang='chi_sim')
        print(text)
        count += 1


if __name__ == '__main__':
    current_dir = os.getcwd()
    print("current_dir:", current_dir)
    # current_dir + "/" + "png"
    # current_dir + "/" + "png"
    # args: ['merge.py', 'png', 'png']

    args = sys.argv
    print('args:', args)
    input_dir = args[1]
    output_dir = args[2]
    # extract_subtitles(input_dir, output_dir)
    # extract_subtitles_v1(input_dir, output_dir)
    extract_subtitles_v3(input_dir, output_dir)