Merge pull request #391 from DoubleStarK/feature/cli

Feature/cli

cli.ini

@@ -6,18 +6,22 @@ proxy=
 ;输出结果文件到目录
 target_dir=
 ;视频发音语言，从这里选择 zh-cn zh-tw en fr de ja ko ru es th it pt vi ar tr
-source_language=zh-cn
+source_language=en
 ;语音识别语言 无需填写
 detect_language=
 ;翻译到的语言 zh-cn zh-tw en fr de ja ko ru es th it pt vi ar tr
-target_language=en
+target_language=zh-cn
 ;软字幕嵌入时的语言，不填写
 subtitle_language=
 ;true=启用CUDA
 cuda=false
-;角色名称，openaiTTS角色名称“alloy,echo,fable,onyx,nova,shimmer”，edgeTTS角色名称从 voice_list.json 中对应语言的角色中寻找。elevenlabsTTS 的角色名称从 elevenlabs.json 中寻找
-voice_role=en-CA-ClaraNeural
-; 配音加速值，必须以 + 号或 - 号开头，+代表加速，-代表减速，以%结尾
+;角色名称(No代表用原音)，openaiTTS角色名称“alloy,echo,fable,onyx,nova,shimmer”，edgeTTS角色名称从 voice_list.json 中对应语言的角色中寻找。elevenlabsTTS 的角色名称从 elevenlabs.json 中寻找
+voice_role=No
+;背景声音音量降低或升高幅度，大于1升高，小于1降低
+volume=+0.5%
+;背景声音变调，大于1升高，小于1降低
+pitch=+0%
+;配音加速值，必须以 + 号或 - 号开头，+代表加速，-代表减速，以%结尾
 voice_rate=+0%
 ;可选 edgetTTS openaiTTS elevenlabsTTS
 tts_type=edgeTTS
@@ -27,8 +31,8 @@ voice_silence=500
 is_separate=false
 ;all=整体识别，split=预先分割声音片段后识别
 whisper_type=all
-;语音识别模型可选，base small medium large-v3
-whisper_model=base
+;语音识别模型可选，tiny base small medium large-v3
+whisper_model=tiny
 model_type=faster
 ;翻译渠道，可选 google baidu chatGPT Azure Gemini tencent DeepL DeepLX
 translate_type=google

cli.py

@@ -11,38 +11,37 @@
 import os
 import re
 import sys
+import traceback
+from videotrans.box.worker import Worker
 from videotrans.configure import config
 import argparse
 from videotrans.task.trans_create import TransCreate
 from videotrans.translator import LANG_CODE, is_allow_translate, BAIDU_NAME, TENCENT_NAME, CHATGPT_NAME, AZUREGPT_NAME, \
  GEMINI_NAME, DEEPLX_NAME, DEEPL_NAME
-from videotrans.util.tools import set_proxy, get_edge_rolelist, get_elevenlabs_role
+from videotrans.util import tools
+from videotrans.util.tools import send_notification, set_process, set_proxy, get_edge_rolelist, get_elevenlabs_role
+from tqdm import tqdm
 
-parser = argparse.ArgumentParser(description='cli.ini and source mp4')
-parser.add_argument('-c', type=str, help='cli.ini file absolute filepath', default=os.path.join(os.getcwd(), 'cli.ini'))
-parser.add_argument('-m', type=str, help='mp4 absolute filepath', default="")
-parser.add_argument('-cuda', action='store_true', help='Activates the cuda option')
+def __init__():
+ if not os.path.exists(os.path.join(config.rootdir, 'voice_list.json')) or os.path.getsize(
+ os.path.join(config.rootdir, 'voice_list.json')) == 0:
+ print("正在获取 edge TTS 角色...")
+ get_edge_rolelist()
+ if not os.path.exists(os.path.join(config.rootdir, 'elevenlabs.json')) or os.path.getsize(
+ os.path.join(config.rootdir, 'elevenlabs.json')) == 0:
+ print("正在获取 elevenlabs TTS 角色...")
+ get_elevenlabs_role()
 
 
-def set_process(text, type="logs"):
- print(f'[{type}] {text}\n')
-
-
-if not os.path.exists(os.path.join(config.rootdir, 'voice_list.json')) or os.path.getsize(
- os.path.join(config.rootdir, 'voice_list.json')) == 0:
- print("正在获取 edge TTS 角色...")
- get_edge_rolelist()
-if not os.path.exists(os.path.join(config.rootdir, 'elevenlabs.json')) or os.path.getsize(
- os.path.join(config.rootdir, 'elevenlabs.json')) == 0:
- print("正在获取 elevenlabs TTS 角色...")
- get_elevenlabs_role()
-
 if __name__ == '__main__':
- config.exec_mode = 'cli'
- config.settings['countdown_sec'] = 0
+ parser = argparse.ArgumentParser(description='cli.ini and source mp4')
+ parser.add_argument('-c', type=str, help='cli.ini file absolute filepath', default=os.path.join(os.getcwd(), 'cli.ini'))
+ parser.add_argument('-m', type=str, help='mp4 absolute filepath', default="")
+ parser.add_argument('-cuda', action='store_true', help='Activates the cuda option')
 
  args = vars(parser.parse_args())
 
+ config.settings['countdown_sec'] = 0
  cfg_file = args['c']
  if not os.path.exists(cfg_file):
  print('不存在配置文件 cli.ini' if config.defaulelang == 'zh' else "cli.ini file not exists")
@@ -67,10 +66,21 @@ def set_process(text, type="logs"):
  config.params['source_language']='-'
  if not config.params['target_language']:
  config.params['target_language']='-'
+ if not config.params.get('back_audio'):
+ config.params['back_audio']='-'
  if args['cuda']:
  config.params['cuda'] = True
  if args['m'] and os.path.exists(args['m']):
  config.params['source_mp4'] = args['m']
+
+ # 传多个视频的话,考虑支持批量处理
+ if type(args['m']) == str:
+ config.params['is_batch'] = False
+ else:
+ config.params['is_batch'] = True
+ print("命令行批量处理暂未支持")
+ sys.exit()
+
  if not config.params['source_mp4'] or not os.path.exists(config.params['source_mp4']):
  print(
  "必须在命令行或cli.ini文件设置 source_mp4(视频文件)的绝对路径" if config.defaulelang == 'zh' else "The absolute path of source_mp4 (video file) must be set on the command line or in the cli.ini file.")
@@ -88,7 +98,7 @@ def set_process(text, type="logs"):
  print(config.transobj['anerror'], config.transobj['baikeymust'])
  sys.exit()
  elif config.params['translate_type'] == TENCENT_NAME:
- #  腾讯翻译
+ # 腾讯翻译
  if not config.params["tencent_SecretId"] or not config.params["tencent_SecretKey"]:
  print(config.transobj['tencent_key'])
  sys.exit()
@@ -135,11 +145,73 @@ def set_process(text, type="logs"):
  config.proxy = config.params['proxy'].strip()
  set_proxy(config.proxy)
  config.current_status = 'ing'
+ config.params['app_mode'] = 'cli'
+ (base, ext) = os.path.splitext(config.params['source_mp4'].replace('\\', '/'))
+ config.params['target_dir'] = os.path.dirname(base)
+ obj_format = tools.format_video(config.params['source_mp4'].replace('\\', '/'), config.params['target_dir'])
+
+ process_bar_data = [
+ config.transobj['kaishichuli'],
+ config.transobj['kaishishibie'],
+ config.transobj['starttrans'],
+ config.transobj['kaishipeiyin'],
+ config.transobj['kaishihebing'],
+ ]
+
+ process_bar = tqdm(process_bar_data)
+ try:
+ video_task = TransCreate(config.params, obj_format)
+ try:
+ process_bar.set_description(process_bar_data[0])
+ video_task.prepare()
+ process_bar.update(1)
+ except Exception as e:
+ err=f'{config.transobj["yuchulichucuo"]}:' + str(e)
+ print(err)
+ sys.exit()
+ try:
+ process_bar.set_description(process_bar_data[1])
+ video_task.recogn()
+ process_bar.update(1)
+ except Exception as e:
+ err=f'{config.transobj["shibiechucuo"]}:' + str(e)
+ print(err)
+ sys.exit()
+ try:
+ process_bar.set_description(process_bar_data[2])
+ video_task.trans()
+ process_bar.update(1)
+ except Exception as e:
+ err=f'{config.transobj["fanyichucuo"]}:' + str(e)
+ print(err)
+ sys.exit()
+ try:
+ process_bar.set_description(process_bar_data[3])
+ video_task.dubbing()
+ process_bar.update(1)
+ except Exception as e:
+ err=f'{config.transobj["peiyinchucuo"]}:' + str(e)
+ print(err)
+ sys.exit()
+ try:
+ process_bar.set_description(process_bar_data[4])
+ video_task.hebing()
+ process_bar.update(1)
+ except Exception as e:
+ err=f'{config.transobj["hebingchucuo"]}:' + str(e)
+ print(err)
+ sys.exit()
+ try:
+ video_task.move_at_end()
+ process_bar.update(1)
+ except Exception as e:
+ err=f'{config.transobj["hebingchucuo"]}:' + str(e)
+ print(err)
+ sys.exit()
 
- # try:
- # task = TransCreate({"source_mp4": config.params['source_mp4'], 'app_mode': "biaozhun","mode":"cli"})
- # set_process(config.transobj['kaishichuli'])
- # res = task.run()
- # print(f'{"执行完成" if config.defaulelang == "zh" else "Succeed"} {task.targetdir_mp4}')
- # except Exception as e:
- # print(f'\nException:{str(e)}\n')
+ send_notification(config.transobj["zhixingwc"], f'"subtitles -> audio"')
+ print(f'{"执行完成" if config.defaulelang == "zh" else "Succeed"} {video_task.targetdir_mp4}')
+ except Exception as e:
+ send_notification(e, f'{video_task.obj["raw_basename"]}')
+ # 捕获异常并重新绑定回溯信息
+ traceback.print_exc()

videotrans/configure/config.py

@@ -140,9 +140,9 @@ def parse_init():
 # 软件界面
 uilanglist = obj["ui_lang"]
 # 语言代码:语言显示名称
-langlist = obj["language_code_list"]
+langlist: dict = obj["language_code_list"]
 # 语言显示名称：语言代码
-rev_langlist = {val: key for key, val in langlist.items()}
+rev_langlist = {code_alias: code for code, code_alias in langlist.items()}
 # 语言显示名称 list
 langnamelist = list(langlist.values())
 # 工具箱语言

videotrans/language/en.json

@@ -13,6 +13,7 @@
  "shibiechucuo": "Error in speech recognition",
  "fanyichucuo": "Error in subtitle translation",
  "peiyinchucuo": "Error in dubbing",
+ "hebingchucuo": "Error in combine",
 
  "freechatgpt_tips": "apiskey.top sponsored ChatGPT, free to use",
 

videotrans/language/es.json

@@ -14,6 +14,7 @@
  "shibiechucuo": "Error in speech recognition",
  "fanyichucuo": "Error in subtitle translation",
  "peiyinchucuo": "Error in dubbing",
+ "hebingchucuo": "Error in combine",
 
  "freechatgpt_tips": "apiskey.top sponsored ChatGPT, free to use",
 

videotrans/set.ini

@@ -8,7 +8,7 @@
 ;Interface language text #############################
 ;默认界面跟随系统，也可以在此手动指定，zh=中文界面，en=英文界面
 ;Default interface follows the system, you can also specify it manually here, zh=Chinese interface, en=English interface.
-lang =
+lang = en
 
 ;##################视频质量############################
 ;Video quality ############################

videotrans/task/trans_create.py

@@ -46,7 +46,7 @@ class TransCreate():
 
  '''
 
- def __init__(self, config_params=None, obj=None):
+ def __init__(self, config_params: dict = None, obj=None):
  # 视频原始路径 名称等信息
  self.obj = obj
  # 配置信息
@@ -164,8 +164,14 @@ def __init__(self, config_params=None, obj=None):
  self.init['target_language_code'] = self.config_params['target_language']
  else:
  # 仅作为文件名标识
- self.init['source_language_code']=config.rev_langlist[self.config_params['source_language']] if self.config_params['source_language'] != '-' else '-'
- self.init['target_language_code'] = config.rev_langlist[self.config_params['target_language']] if self.config_params['target_language'] != '-' else '-'
+ var_a = config.rev_langlist.get(self.config_params['source_language'])
+ var_b = config.langlist.get(self.config_params['source_language'])
+ var_c = var_a if var_a is not None else var_b
+ self.init['source_language_code'] = var_c if var_c != '-' else '-'
+ var_a = config.rev_langlist.get(self.config_params['target_language'])
+ var_b = config.langlist.get(self.config_params['target_language'])
+ var_c = var_a if var_a is not None else var_b
+ self.init['target_language_code'] = var_c if var_c != '-' else '-'
 
  # 检测字幕原始语言
  if self.config_params['source_language'] != '-':
@@ -248,6 +254,7 @@ def runing():
 
  self._split_wav_novicemp4()
  self.step_inst=Runstep(init=self.init,obj=self.obj,config_params=self.config_params,parent=self)
+ return True
 
  def __getattr__(self, precent):
  return self.step_inst.precent if self.step_inst else 0
@@ -421,9 +428,11 @@ def move_at_end(self):
  btnkey=self.init['btnkey']
  )
  tools.send_notification("Succeed", f"{self.obj['raw_basename']}")
+
  #删除临时文件
  shutil.rmtree(self.init['cache_folder'], ignore_errors=True)
  if linshi_deldir:
  shutil.rmtree(linshi_deldir)
  if wait_deldir:
  shutil.rmtree(wait_deldir,ignore_errors=True)
+ return True

videotrans/tts/__init__.py

@@ -62,7 +62,7 @@ def text_to_speech(
  config.logger.error(f'no filename={filename} {tts_type=} {text=},{role=}')
 
 
-def run(*, queue_tts=None, language=None,set_p=True,inst=None):
+def run(*, queue_tts=None, language=None, set_p=True, inst=None):
  queue_tts_copy=copy.deepcopy(queue_tts)
  # 需要并行的数量3
  n_total = len(queue_tts)