.srt ファイルを日本語翻訳するツール。注意：節度をもって使うこと。google translate にChrome(selenium webdriver)の自動操作でアクセスします。

translate_srt.py

# -*- coding:utf-8 -*-
from selenium import webdriver

user_agent = (
    "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_8_4) " +
    "AppleWebKit/537.36 (KHTML, like Gecko) Chrome/29.0.1547.57 Safari/537.36"
)

sourcedir = "/tmp"
sourcefiles = ["subfile1.srt", "subfile2.srt"]


import time

class GoogletTanslate(object):
  driver = None
  chromedriver = "/Users/t2psyto/Downloads/chromedriver"
  phantomjs = "/Users/t2psyto/bin/phantomjs"
  def create_driver(self):
    if self.driver == None:
      self.driver = webdriver.Chrome(executable_path=self.chromedriver)
      #self.driver = driver = webdriver.PhantomJS(executable_path=self.phantomjs, desired_capabilities=dcap, service_args=['--ignore-ssl-errors=true'])
    return self.driver
      
  def open_google_translate(self):
    self.create_driver()
    self.driver.get("https://translate.google.com/?hl=ja")

  def translate(self, inputtext):
    driver = self.driver

    lang_from = driver.find_element_by_xpath('//*[@id="gt-sl-sugg"]/div[1]')
    lang_from.click()
    lang_to = driver.find_element_by_xpath('//*[@id="gt-tl-sugg"]/div[1]')
    lang_to.click()

    #paste src
    elem = driver.find_element_by_xpath('//*[@id="source"]')
    driver.execute_script("arguments[0].value=arguments[1]", elem, inputtext)
    
    time.sleep(2)
    #print elem.text

    #honyaku
    form_submit = driver.find_element_by_xpath('//*[@id="gt-lang-submit"]')
    form_submit.click()

    time.sleep(3)

    #get translated text
    result_box = driver.find_element_by_xpath('//*[@id="result_box"]')
    result_text = result_box.text

    return result_text

if __name__ == '__main__':

    import pysrt

    
    gt = GoogletTanslate()
    gt.open_google_translate()


    for sourcefile in sourcefiles:
        print sourcefile, ": open."
        sourcefullpath = sourcedir + "/" + sourcefile
        subs = pysrt.open(sourcefullpath)

        suben = []
        for sub in subs:
            #suben.append(u"#%d: %s" % (sub.index, sub.text.replace("\n", " ")) )
            stren = sub.text.replace("\n", " ")
            #行数を揃えるため、空行を適当な文字で埋めておく。
            if stren.strip() == "":
              stren = "-"
            suben.append(stren)

        srten = u"\n".join(suben)
        print sourcefile, ": translate begin."
        retja = gt.translate(srten)
        print sourcefile, ": translate done."

        #lija = retja.replace("＃".decode("utf-8"),u"").replace("：".decode("utf-8"),u"\t").split(u"\n")
        lija = retja.split(u"\n")

        for i,sub in enumerate(subs):
            subtexten = suben[i]
            #subtextja = lija[i].split(u"\t")[1]
            subtextja = lija[i]
            #字幕の 1行目 英語、2行目 日本語
            sub.text = subtexten + "\n" + subtextja

        subs.save(sourcedir + "/" + "ja_" + sourcefile, "utf-8")
        print sourcefile, ": saved."


    gt.driver.close()