add microsoft translator

jianchang512 · Feb 8, 2024 · 88f9b28 · 88f9b28
1 parent b355762
commit 88f9b28
Show file tree

Hide file tree

Showing 16 changed files with 194 additions and 94 deletions.
diff --git a/test.py b/test.py
@@ -33,49 +33,38 @@
 #     msg=f"separate vocal and background music:{str(e)}"
 #     #set_process(msg)
 #     print(msg)
-import sys
 
-from PySide6.QtGui import QGuiApplication
-from PySide6.QtWidgets import QWidget, QApplication
 
-def ceshi():
-    from you_get.extractors import youtube
+## 获取识别文字
+# import os
+# from faster_whisper import WhisperModel
+# model = WhisperModel("base", device="cpu",
+#                              download_root="./models",
+#                              local_files_only=True)
+# data=[]
+#
+# for i in range(0,24):
+#     name=f'output0{str(i).zfill(2)}.wav'
+#     if not os.path.exists(f'c:/users/c1/videos/_video_out/{name}'):
+#         continue
+#     segments, info = model.transcribe(f'c:/users/c1/videos/_video_out/{name}',
+#                                           beam_size=1,
+#                                           best_of=1,
+#                                           condition_on_previous_text=False,language="zh")
+#     res=[]
+#     for segment in segments:
+#         res.append(segment.text.strip())
+#     data.append(f"wavs/{name}|{'.'.join(res)}|coqui")
+#
+# with open("./metadata_train.csv","w",encoding='utf-8') as f:
+#     f.write("\n".join(data))
 
-    try:
-        youtube.download('https://www.youtube.com/watch?v=n4mF5xo5khM',
-                         output_dir='.',
-                         merge=True,
-                         extractor_proxy='http://127.0.0.1:10809'
-                         )
-    except Exception as e:
-        print(e)
-
-class StartWindow(QWidget):
-    def __init__(self):
-        super(StartWindow, self).__init__()
-        # 设置窗口无边框和标题
-        # 设置窗口的背景图片
-        # 窗口大小
-        self.resize(560, 350)
-        self.center()
-        self.show()
-        # 使用QTimer延时显示窗口B
-        ceshi()
-
-
-    def center(self):
-        screen=QGuiApplication.primaryScreen()
-        qtRect = self.frameGeometry()
-        qtRect.moveCenter(screen.availableGeometry().center())
-        reso=screen.geometry()
-        self.width, self.height=reso.width(), reso.height()
-        self.move(qtRect.topLeft())
+import requests
+lang=["zh-hans","zh-hant","en","fr","de","ja","ko","ru","es","th","it","pt","vi","ar","tr","hi","hu"]
 
-if __name__ == "__main__":
-    app = QApplication(sys.argv)
-    try:
-        startwin = StartWindow()
-    except Exception as e:
-        print(f"error:{str(e)}")
-    sys.exit(app.exec())
+auth=requests.get('https://edge.microsoft.com/translate/auth')
 
+for it in lang:
+    url=f'https://api-edge.cognitive.microsofttranslator.com/translate?from=&to={it}&api-version=3.0&includeSentenceLength=true'
+    res=requests.post(url,json=[{"Text":"hello,my friend\nI am from China"}],headers={"Authorization":f"Bearer {auth.text}"})
+    print(res.json())
diff --git a/version.json b/version.json
@@ -1,4 +1,4 @@
 {
-  "version": "0.9997",
-  "version_num": 10007
+  "version": "0.9998",
+  "version_num": 10008
 }
diff --git a/videotrans/__init__.py b/videotrans/__init__.py
@@ -1,4 +1,4 @@
 # -*- coding: utf-8 -*-
 
-VERSION="0.9997 v.wonyes.org"
-VERSION_NUM=10007
+VERSION="0.9998 v.wonyes.org"
+VERSION_NUM=10008
diff --git a/videotrans/language/en.json b/videotrans/language/en.json
@@ -1,5 +1,6 @@
 {
   "translate_language": {
+    "test google": "Test connection Google..",
     "Select Out Dir": "Select Out Dir",
     "downing...": "Downloading...",
     "start download": "Start Download",

diff --git a/videotrans/language/es.json b/videotrans/language/es.json
@@ -1,5 +1,6 @@
 {
   "translate_language" : {
+    "test google": "Test connection Google..",
     "Select Out Dir": "Select Out Dir",
     "downing...": "Downloading...",
     "start download": "Start Download",

diff --git a/videotrans/language/zh.json b/videotrans/language/zh.json
@@ -1,5 +1,6 @@
 {
   "translate_language": {
+    "test google": "正在测试能否连接到Google..",
     "Select Out Dir": "下载到..",
     "downing...": "下载中...",
     "start download": "立即下载",
@@ -179,7 +180,7 @@
     "Open": "打开",
     "Translate channel": "翻译渠道",
     "Proxy": "代理地址",
-    "proxy address": "Google或chatGPT官方接口,国内必填代理或设置系统代理,格式http://127.0.0.1:端口号",
+    "proxy address": "Google或chatGPT官方接口需填代理,代理格式 http://127.0.0.1:端口号 ",
     "shuoming01": "点击试听当前配音角色的发音\n生成配音可能需要数秒，请耐心等待",
     "Trial dubbing": "试听配音",
     "Source lang": "原始语言",
@@ -224,17 +225,17 @@
     "Donating developers": "捐助开发者",
     "Standard Function Mode": "标准功能模式",
     "Display all options for video translation and dubbing": "显示全部选项，进行视频翻译和配音",
-    "Export  Srt  From Videos": "视频提取字幕",
+    "Export  Srt  From Videos": "视频识别字幕",
     "Extracting SRT subtitles in the original language from local videos": "从本地视频中提取出原始语言的srt字幕",
     "Merging Subtitle  Video": "字幕视频合并",
     "Embed locally existing SRT subtitles into the video": "将本地已有的srt字幕嵌入视频中",
     "Subtitle Create Dubbing": "字幕创建配音",
     "Local existing SRT subtitle generation dubbing WAV files": "本地已有的srt字幕生成配音wav文件",
-    "Speech Recognition Text": "语音识别文字",
+    "Speech Recognition Text": "音视频转文字",
     "Recognize the sound in audio or video and output SRT text": "将音频或视频中的声音识别后输出srt文字",
     "From  Text  Into  Speech": "文字合成语音",
     "Generate audio WAV from text or SRT subtitle files": "将文字或srt字幕文件生成音频wav",
-    "Extract Srt And Translate": "提取字幕翻译",
+    "Extract Srt And Translate": "识别字幕翻译",
     "Extract SRT subtitles from local videos in the original language and translate them into SRT subtitle files in the target language": "从本地视频中提取出原始语言的srt字幕,并翻译为目标语言的srt字幕文件",
     "Separate Video to audio": "视频音频分离",
     "Separate audio and silent videos from videos": "从视频中分离出音频和无声视频",

diff --git a/videotrans/task/trans_create.py b/videotrans/task/trans_create.py
@@ -160,6 +160,7 @@ def __init__(self, obj):
             if os.path.exists(self.targetdir_source_regcon) and os.path.getsize(self.targetdir_source_regcon)==0:
                 os.unlink(self.targetdir_source_regcon)
         else:
+            self.targetdir_source_vocal=self.targetdir_source_wav
             self.source_separate = self.source_back = self.source_vocal = None
 
         # 如果存在字幕，则视为目标字幕，直接生成，不再识别和翻译
@@ -315,11 +316,14 @@ def trans(self):
         if not os.path.exists(self.targetdir_source_sub) or os.path.exists(self.targetdir_target_sub):
             return True
         # 测试翻译
+        switch_trans=""
         if config.params['translate_type'].lower()==GOOGLE_NAME.lower():
-            set_process("Test Google connecting...")
+            set_process(config.transobj['test google'])
             if not self.testgoogle():
-                raise Exception(f'无法连接Google，请正确设置代理，例如v2ray默认是http://127.0.0.1:10809,clash默认http://127.0.0.1:7890或开启全局代理')
-        set_process(transobj['starttrans'])
+                switch_trans='无法连接Google，已自动切换为Microsoft翻译'
+                config.params['translate_type']='Microsoft'
+
+        set_process(transobj['starttrans']+switch_trans)
         # 开始翻译,从目标文件夹读取原始字幕
         rawsrt = get_subtitle_from_srt(self.targetdir_source_sub, is_file=True)
         if not rawsrt or len(rawsrt)<1:
@@ -496,11 +500,11 @@ def before_tts(self):
                 filename = self.cache_folder + "/" + md5_hash.hexdigest() + ".mp3"
                 # 如果是clone-voice类型， 需要截取对应片段
                 if config.params['tts_type']=='clone-voice':
-                    if not os.path.exists(self.targetdir_source_vocal):
+                    if config.params['is_separate'] and not os.path.exists(self.targetdir_source_vocal):
                         raise Exception(f'not exits {self.targetdir_source_vocal}')
-                    # clone 方式文件为wav格式
+                        # clone 方式文件为wav格式
                     filename+=".wav"
-                    cut_from_audio(audio_file=self.targetdir_source_vocal,ss=it['startraw'],to=it['endraw'],out_file=filename)
+                    cut_from_audio(audio_file=self.targetdir_source_vocal if config.params['is_separate'] else self.targetdir_source_wav,ss=it['startraw'],to=it['endraw'],out_file=filename)
 
                 queue_tts.append({
                     "text": it['text'],

diff --git a/videotrans/translator/__init__.py b/videotrans/translator/__init__.py
@@ -3,6 +3,7 @@
 from videotrans.configure import config
 
 GOOGLE_NAME = "Google"
+MICROSOFT_NAME = "Microsoft"
 BAIDU_NAME = "Baidu"
 DEEPL_NAME = "DeepL"
 DEEPLX_NAME = "DeepLx"
@@ -14,6 +15,7 @@
 SRT_NAME = "srt"
 # 翻译通道
 TRANSNAMES = [
+    MICROSOFT_NAME,
     GOOGLE_NAME,
     BAIDU_NAME,
     DEEPL_NAME,
@@ -32,22 +34,25 @@
         "zh",  # 百度通道
         "ZH",  # deepl deeplx通道
         "zh",  # 腾讯通道
-        "zh"  # OTT通道
+        "zh",  # OTT通道
+        "zh-Hans"
     ],
     "zh-tw": [
         "zh-tw",
         "chi",
         "cht",
         "ZH",
         "zh-TW",
-        "zt"
+        "zt",
+        "zh-Hant"
     ],
     "en": [
         "en",
         "eng",
         "en",
         "EN",
         "en",
+        "en",
         "en"
     ],
     "fr": [
@@ -56,6 +61,7 @@
         "fra",
         "FR",
         "fr",
+        "fr",
         "fr"
     ],
     "de": [
@@ -64,6 +70,7 @@
         "de",
         "DE",
         "de",
+        "de",
         "de"
     ],
     "ja": [
@@ -72,6 +79,7 @@
         "jp",
         "JA",
         "ja",
+        "ja",
         "ja"
     ],
     "ko": [
@@ -80,6 +88,7 @@
         "kor",
         "KO",
         "ko",
+        "ko",
         "ko"
     ],
     "ru": [
@@ -88,6 +97,7 @@
         "ru",
         "RU",
         "ru",
+        "ru",
         "ru"
     ],
     "es": [
@@ -96,6 +106,7 @@
         "spa",
         "ES",
         "es",
+        "es",
         "es"
     ],
     "th": [
@@ -104,6 +115,7 @@
         "th",
         "No",
         "th",
+        "th",
         "th"
     ],
     "it": [
@@ -112,6 +124,7 @@
         "it",
         "IT",
         "it",
+        "it",
         "it"
     ],
     "pt": [
@@ -120,6 +133,7 @@
         "pt",
         "PT",
         "pt",
+        "pt",
         "pt"
     ],
     "vi": [
@@ -128,14 +142,16 @@
         "vie",
         "No",
         "vi",
-        "No"
+        "No",
+        "vi"
     ],
     "ar": [
         "ar",
         "are",
         "ara",
         "No",
         "ar",
+        "ar",
         "ar"
     ],
     "tr": [
@@ -144,6 +160,7 @@
         "tr",
         "tr",
         "tr",
+        "tr",
         "tr"
     ],
     "hi": [
@@ -152,6 +169,7 @@
         "hi",
         "No",
         "hi",
+        "hi",
         "hi"
     ],
     "hu": [
@@ -160,7 +178,8 @@
         "hu",
         "HU",
         "No",
-        "No"
+        "No",
+        "hu"
     ]
 }
 
@@ -183,7 +202,6 @@ def get_source_target_code(*, show_source=None, show_target=None, translate_type
     if not translate_type:
         return None, None
     lower_translate_type = translate_type.lower()
-    print(f'{lower_translate_type == GOOGLE_NAME.lower()}')
     if show_source:
         source_list = LANG_CODE[show_source] if show_source in LANG_CODE else LANG_CODE[
             config.rev_langlist[show_source]]
@@ -202,6 +220,8 @@ def get_source_target_code(*, show_source=None, show_target=None, translate_type
         return (show_source, show_target)
     elif lower_translate_type == OTT_NAME.lower():
         return (source_list[5] if source_list else "-", target_list[5] if target_list else "-")
+    elif lower_translate_type==MICROSOFT_NAME.lower():
+        return (source_list[6] if source_list else "-", target_list[6] if target_list else "-")
     else:
         raise Exception(f"[error]get_source_target_code:{translate_type=},{show_source=},{show_target=}")
 
@@ -237,14 +257,16 @@ def is_allow_translate(*, translate_type=None, show_target=None, only_key=False)
 
     if only_key:
         return True
-
+    #再判断是否为No，即不支持
     index = 0
     if lower_translate_type == BAIDU_NAME.lower():
         index = 2
     elif lower_translate_type in [DEEPLX_NAME.lower(), DEEPL_NAME.lower()]:
         index = 3
     elif lower_translate_type == TENCENT_NAME.lower():
         index = 4
+    elif lower_translate_type==MICROSOFT_NAME.lower():
+        index=6
 
     if show_target:
         target_list = LANG_CODE[show_target] if show_target in LANG_CODE else LANG_CODE[
@@ -290,6 +312,8 @@ def run(*, translate_type=None, text_list=None, target_language_name=None, set_p
         from videotrans.translator.gemini import trans
     elif lower_translate_type == AZUREGPT_NAME.lower():
         from videotrans.translator.azure import trans
+    elif lower_translate_type==MICROSOFT_NAME.lower():
+        from videotrans.translator.microsoft import trans
     else:
         print(lower_translate_type == OTT_NAME.lower())
         raise Exception(f"[error]run {translate_type=},{target_language_name=}")