fix: bugs and update

videotrans/__init__.py

@@ -1,4 +1,4 @@
 # -*- coding: utf-8 -*-
 
-VERSION="v1.81"
-VERSION_NUM=110781
+VERSION="v1.82"
+VERSION_NUM=110782

videotrans/box/win.py

@@ -129,36 +129,7 @@ def initUI(self):
         self.tts_type.currentTextChanged.connect(self.tts_type_change)
         self.tts_issrt.stateChanged.connect(self.tts_issrt_change)
 
-        # tab-5 格式转换
-        # self.geshi_input = TextGetdir()
-        # sizePolicy = QtWidgets.QSizePolicy(QtWidgets.QSizePolicy.Expanding, QtWidgets.QSizePolicy.Expanding)
-        # sizePolicy.setHorizontalStretch(0)
-        # sizePolicy.setVerticalStretch(0)
-        # sizePolicy.setHeightForWidth(self.geshi_result.sizePolicy().hasHeightForWidth())
-        # self.geshi_input.setSizePolicy(sizePolicy)
-        # self.geshi_input.setMinimumSize(300, 0)
-        #
-        # self.geshi_input.setPlaceholderText(config.transobj['tuodongdaoci'])
-        #
-        # self.geshi_importbtn = QtWidgets.QPushButton(self.tab_6)
-        # self.geshi_importbtn.setObjectName("geshi_importbtn")
-        # self.geshi_importbtn.setFixedWidth(100)
-        # self.geshi_importbtn.setText(config.box_lang['import audio or video'])
-        # self.geshi_importbtn.clicked.connect(lambda: self.geshi_import_fun(self.geshi_input))
-        # self.horizontalLayout_14.insertWidget(0, self.geshi_importbtn)
-        #
-        # self.geshi_layout.insertWidget(0, self.geshi_input)
-        # self.geshi_mp4.clicked.connect(lambda: self.geshi_start_fun("mp4"))
-        # self.geshi_avi.clicked.connect(lambda: self.geshi_start_fun("avi"))
-        # self.geshi_mov.clicked.connect(lambda: self.geshi_start_fun("mov"))
-        # self.geshi_mp3.clicked.connect(lambda: self.geshi_start_fun("mp3"))
-        # self.geshi_wav.clicked.connect(lambda: self.geshi_start_fun("wav"))
-        # self.geshi_aac.clicked.connect(lambda: self.geshi_start_fun("aac"))
-        # self.geshi_m4a.clicked.connect(lambda: self.geshi_start_fun("m4a"))
-        # self.geshi_flac.clicked.connect(lambda: self.geshi_start_fun("flac"))
-        # self.geshi_output.clicked.connect(lambda: self.opendir_fn(f'{config.homedir}/conver'))
-        # if not os.path.exists(f'{config.homedir}/conver'):
-        #     os.makedirs(f'{config.homedir}/conver', exist_ok=True)
+
 
         # 混流
         self.hun_file1btn.clicked.connect(lambda: self.hun_get_file('file1'))
@@ -698,7 +669,7 @@ def listen_voice_fun(self):
 
 
         voice_file = f"{voice_dir}/{config.params['tts_type']}-{lang}-{lujing_role}-{volume}-{pitch}.mp3"
-        if config.params['tts_type'] == 'GPT-SoVITS':
+        if config.params['tts_type'] in ['GPT-SoVITS','ChatTTS']:
             voice_file += '.wav'
 
         obj = {
@@ -753,6 +724,13 @@ def hecheng_start_fun(self):
             self.tts_type.setCurrentText('edgeTTS')
             QMessageBox.critical(self, config.transobj['anerror'], config.transobj['nogptsovitslanguage'])
             return
+        if tts_type == 'ChatTTS' and langcode[:2] not in ['zh', 'en']:
+            # 除此指望不支持
+            config.params['tts_type'] = 'edgeTTS'
+            self.tts_type.setCurrentText('edgeTTS')
+            QMessageBox.critical(self, config.transobj['anerror'], config.transobj['onlycnanden'])
+            return
+
         if rate >= 0:
             rate = f"+{rate}%"
         else:
@@ -822,10 +800,17 @@ def tts_type_change(self, type):
             self.volume_rate.setDisabled(True)
             self.pitch_rate.setDisabled(True)
 
-
+        code = translator.get_code(show_text=self.hecheng_language.currentText())
         if type == 'gtts':
             self.hecheng_role.clear()
             self.hecheng_role.addItems(['gtts'])
+        elif type=='ChatTTS':
+            if code and code != '-' and code[:2] not in ['zh',  'en']:
+                self.tts_type.setCurrentText('edgeTTS')
+                QMessageBox.critical(self, config.transobj['anerror'], config.transobj['onlycnanden'])
+                return
+            self.hecheng_role.clear()
+            self.hecheng_role.addItems(['No']+list(config.ChatTTS_voicelist.keys()) )
         elif type == "openaiTTS":
             self.hecheng_role.clear()
             self.hecheng_role.addItems(config.params['openaitts_role'].split(","))
@@ -844,7 +829,6 @@ def tts_type_change(self, type):
             self.hecheng_role.clear()
             self.hecheng_role.addItems(config.params['ttsapi_voice_role'].split(","))
         elif type == 'GPT-SoVITS':
-            code = translator.get_code(show_text=self.hecheng_language.currentText())
             if code and code != '-' and code[:2] not in ['zh', 'ja', 'en']:
                 self.tts_type.setCurrentText('edgeTTS')
                 QMessageBox.critical(self, config.transobj['anerror'], config.transobj['nogptsovitslanguage'])
@@ -860,6 +844,13 @@ def hecheng_language_fun(self, t):
             # 除此指望不支持
             QMessageBox.critical(self, config.transobj['anerror'], config.transobj['nogptsovitslanguage'])
             self.tts_type.setCurrentText('edgeTTS')
+            return
+        if code and code != '-' and self.tts_type.currentText() == 'ChatTTS' and code[:2] not in ['zh', 'en']:
+            self.tts_type.setCurrentText('edgeTTS')
+            # 除此指望不支持
+            QMessageBox.critical(self, config.transobj['anerror'], config.transobj['onlycnanden'])
+            return
+
         if self.tts_type.currentText() not in ["edgeTTS", "AzureTTS"]:
             return
         self.hecheng_role.clear()
@@ -908,13 +899,6 @@ def hun_fun(self):
         file1 = self.hun_file1.text()
         file2 = self.hun_file2.text()
 
-        #rs, newfile1, _ = tools.rename_move(file1, is_dir=False)
-        #if rs:
-        #    file1 = newfile1
-        #rs, newfile2, _ = tools.rename_move(file2, is_dir=False)
-        #if rs:
-        #    file2 = newfile2
-
         cmd = ['-y', '-i', file1, '-i', file2, '-filter_complex',
                "[0:a][1:a]amix=inputs=2:duration=first:dropout_transition=2", '-ac', '2', savename]
         self.geshi_task = Worker([cmd], "hunhe", self)

videotrans/component/__init__.py

@@ -1,6 +1,6 @@
 from videotrans.component.set_form import BaiduForm, \
     ChatgptForm, DeepLForm, DeepLXForm, TencentForm, ElevenlabsForm, InfoForm, AzureForm, GeminiForm, SetLineRole, \
-    YoutubeForm, OttForm,CloneForm,SeparateForm,TtsapiForm,GPTSoVITSForm,TransapiForm,ArticleForm,AzurettsForm,ZhrecognForm
+    YoutubeForm, OttForm,CloneForm,SeparateForm,TtsapiForm,GPTSoVITSForm,TransapiForm,ArticleForm,AzurettsForm,ZhrecognForm,ChatttsForm
 
 __all__ = [
     "BaiduForm",
@@ -9,5 +9,5 @@
     "DeepLXForm",
     "TencentForm",
     "ElevenlabsForm",
-    "InfoForm", "AzureForm", "GeminiForm", "SetLineRole", "ElevenlabsForm", "YoutubeForm","OttForm","CloneForm","SeparateForm","TtsapiForm","TransapiForm","ArticleForm","AzurettsForm","ZhrecognForm"
+    "InfoForm", "AzureForm", "GeminiForm", "SetLineRole", "ElevenlabsForm", "YoutubeForm","OttForm","CloneForm","SeparateForm","TtsapiForm","TransapiForm","ArticleForm","AzurettsForm","ZhrecognForm","ChatttsForm"
 ]

videotrans/component/set_form.py

@@ -8,6 +8,7 @@
 from videotrans.ui.azure import Ui_azureform
 from videotrans.ui.baidu import Ui_baiduform
 from videotrans.ui.chatgpt import Ui_chatgptform
+from videotrans.ui.chattts import Ui_chatttsform
 from videotrans.ui.deepl import Ui_deeplform
 from videotrans.ui.deeplx import Ui_deeplxform
 from videotrans.ui.gptsovits import Ui_gptsovitsform
@@ -155,6 +156,12 @@ def __init__(self, parent=None):
         self.setupUi(self)
         self.setWindowFlags(QtCore.Qt.WindowStaysOnTopHint)
         self.setWindowIcon(QIcon(f"{config.rootdir}/videotrans/styles/icon.ico"))
+class ChatttsForm(QDialog, Ui_chatttsform):  # <===
+    def __init__(self, parent=None):
+        super(ChatttsForm, self).__init__(parent)
+        self.setupUi(self)
+        self.setWindowFlags(QtCore.Qt.WindowStaysOnTopHint)
+        self.setWindowIcon(QIcon(f"{config.rootdir}/videotrans/styles/icon.ico"))
 
 class ZhrecognForm(QDialog, Ui_zhrecognform):  # <===
     def __init__(self, parent=None):

videotrans/configure/config.py

@@ -185,6 +185,8 @@ def parse_init():
 
 clone_voicelist=["clone"]
 
+ChatTTS_voicelist={"boy1":2222,"boy2":7869,"boy3":6653,"girl1":4099,"girl2":5099}
+
 openaiTTS_rolelist = "alloy,echo,fable,onyx,nova,shimmer"
 chatgpt_model_list = [ it.strip() for it in settings['chatgpt_model'].split(',')]
 # 存放 edget-tts 角色列表
@@ -227,7 +229,7 @@ def parse_init():
     "listen_text_hu": "Helló kedves barátom. Remélem minden napod szép és kellemes!",
 
     "tts_type": "edgeTTS",  # 所选的tts==edge-tts:openaiTTS|coquiTTS|elevenlabsTTS
-    "tts_type_list": ["edgeTTS","gtts","AzureTTS", "GPT-SoVITS","clone-voice","openaiTTS", "elevenlabsTTS","TTS-API"],
+    "tts_type_list": ["edgeTTS","ChatTTS","gtts","AzureTTS", "GPT-SoVITS","clone-voice","openaiTTS", "elevenlabsTTS","TTS-API"],
 
     "whisper_type": "all",
     "whisper_model": "tiny",

videotrans/language/en.json

@@ -1,5 +1,7 @@
 {
   "translate_language": {
+    "onlycnanden": "ChatTTS only support Chinese and English",
+
     "peiyindayu31": "The number of dubbing errors is greater than 1/3, please check the",
 
 	"chaochu255":"The original video path and name is too long, please shorten the video name and move it to a shorter path to avoid subsequent errors.",  

videotrans/language/es.json

@@ -1,5 +1,7 @@
 {
   "translate_language" : {
+    "onlycnanden": "ChatTTS only support Chinese and English",
+
     "peiyindayu31": "The number of dubbing errors is greater than 1/3, please check the",
 
 	"chaochu255":"The original video path and name is too long, please shorten the video name and move it to a shorter path to avoid subsequent errors.",  

videotrans/language/zh.json

@@ -1,5 +1,7 @@
 {
   "translate_language": {
+    "onlycnanden": "ChatTTS仅支持中文和英文配音",
+
     "peiyindayu31": "配音出错数量大于1/3，请检查",
 
 	"chaochu255":"视频路径加名称过长，请缩短视频名称并移动到简短路径下，避免后续出错",  

videotrans/mainwin/secwin.py

@@ -772,18 +772,26 @@ def tts_type_change(self, type):
             self.main.tts_type.setCurrentText(config.params['tts_type_list'][0])
             self.main.subform.set_gptsovits()
             return
+        if type == 'ChatTTS' and not config.params['chattts_api']:
+            self.main.tts_type.setCurrentText(config.params['tts_type_list'][0])
+            self.main.subform.set_chattts_address()
+            return
+
         lang = translator.get_code(show_text=self.main.target_language.currentText())
         if lang and lang != '-' and type == 'GPT-SoVITS' and lang[:2] not in ['zh', 'ja', 'en']:
             self.main.tts_type.setCurrentText(config.params['tts_type_list'][0])
             QMessageBox.critical(self.main, config.transobj['anerror'], config.transobj['nogptsovitslanguage'])
             return
+        if lang and lang != '-' and type == 'ChatTTS' and lang[:2] not in ['zh', 'en']:
+            self.main.tts_type.setCurrentText(config.params['tts_type_list'][0])
+            QMessageBox.critical(self.main, config.transobj['anerror'], config.transobj['onlycnanden'])
+            return
 
         config.params['tts_type'] = type
         config.params['line_roles'] = {}
         if type=='gtts':
             self.main.voice_role.clear()
             self.main.current_rolelist = ["gtts"]
-
             self.main.voice_role.addItems(self.main.current_rolelist)
         elif type == "openaiTTS":
             self.main.voice_role.clear()
@@ -806,7 +814,10 @@ def tts_type_change(self, type):
             self.main.current_rolelist = config.clone_voicelist
             self.main.voice_role.addItems(self.main.current_rolelist)
             threading.Thread(target=tools.get_clone_role).start()
-
+        elif type =='ChatTTS':
+            self.main.voice_role.clear()
+            self.main.current_rolelist = list(config.ChatTTS_voicelist.keys())
+            self.main.voice_role.addItems(['No']+self.main.current_rolelist)
         elif type == 'TTS-API':
             self.main.voice_role.clear()
             self.main.current_rolelist = config.params['ttsapi_voice_role'].strip().split(',')
@@ -837,7 +848,7 @@ def listen_voice_fun(self):
         volume=int(self.main.volume_rate.value())
         pitch=int(self.main.pitch_rate.value())
         voice_file = f"{voice_dir}/{config.params['tts_type']}-{lang}-{lujing_role}-{volume}-{pitch}.mp3"
-        if config.params['tts_type'] == 'GPT-SoVITS':
+        if config.params['tts_type'] in ['GPT-SoVITS','ChatTTS']:
             voice_file += '.wav'
 
         obj = {
@@ -889,6 +900,11 @@ def set_voice_role(self, t):
             config.params['tts_type'] = 'edgeTTS'
             self.main.tts_type.setCurrentText('edgeTTS')
             return QMessageBox.critical(self.main, config.transobj['anerror'], config.transobj['nogptsovitslanguage'])
+        if code and code != '-' and config.params['tts_type'] == 'ChatTTS' and code[:2] not in ['zh','en']:
+            # 除此指望不支持
+            config.params['tts_type'] = 'edgeTTS'
+            self.main.tts_type.setCurrentText('edgeTTS')
+            return QMessageBox.critical(self.main, config.transobj['anerror'], config.transobj['onlycnanden'])
 
         # 除 edgeTTS外，其他的角色不会随语言变化
         if config.params['tts_type'] not in ['edgeTTS', 'AzureTTS']:
@@ -1162,7 +1178,6 @@ def check_start(self):
 
         try:
             voice_rate = int(self.main.voice_rate.value())
-            # voice_rate = 0 if not voice_rate else int(voice_rate)
             config.params['voice_rate'] = f"+{voice_rate}%" if voice_rate >= 0 else f"{voice_rate}%"
         except Exception:
             config.params['voice_rate'] = '+0%'
@@ -1255,6 +1270,12 @@ def check_start(self):
                 QMessageBox.critical(self.main, config.transobj['anerror'], rs)
                 return False
 
+        if self.main.app_mode !='hebing' and config.params['target_language']!='-':
+            code=translator.get_code(show_text=config.params['target_language'])
+            if code not in ['zh-cn','zh-tw','en'] and config.params['tts_type']=='ChatTTS' and config.params['voice_role']!='No':
+                return QMessageBox.critical(self.main, config.transobj['anerror'], config.transobj['onlycnanden'])
+
+
         # 存在视频
         config.params['only_video'] = False
         if config.params['voice_role'] == 'No':

videotrans/mainwin/spwin.py

@@ -202,6 +202,8 @@ def bind_action(self):
         if config.params['tts_type'] == 'clone-voice':
             self.voice_role.addItems(config.clone_voicelist)
             threading.Thread(target=tools.get_clone_role).start()
+        elif config.params['tts_type']=='ChatTTS':
+            self.voice_role.addItems(['No']+list(config.ChatTTS_voicelist.keys()))
         elif config.params['tts_type'] == 'TTS-API':
             self.voice_role.addItems(config.params['ttsapi_voice_role'].strip().split(','))
         elif config.params['tts_type'] == 'GPT-SoVITS':
@@ -290,6 +292,7 @@ def bind_action(self):
         self.actiondeepLX_address.triggered.connect(self.subform.set_deepLX_address)
         self.actionott_address.triggered.connect(self.subform.set_ott_address)
         self.actionclone_address.triggered.connect(self.subform.set_clone_address)
+        self.actionchattts_address.triggered.connect(self.subform.set_chattts_address)
         self.actiontts_api.triggered.connect(self.subform.set_ttsapi)
         self.actionzhrecogn_api.triggered.connect(self.subform.set_zh_recogn)
         self.actiontrans_api.triggered.connect(self.subform.set_transapi)
@@ -426,6 +429,7 @@ def get_setting(self):
         config.params["deeplx_address"] = self.settings.value("deeplx_address", "")
         config.params["ott_address"] = self.settings.value("ott_address", "")
         config.params["clone_api"] = self.settings.value("clone_api", "")
+        config.params["chattts_api"] = self.settings.value("chattts_api", "")
         config.params["tencent_SecretId"] = self.settings.value("tencent_SecretId", "")
         config.params["tencent_SecretKey"] = self.settings.value("tencent_SecretKey", "")
 
@@ -497,6 +501,7 @@ def save_setting(self):
         self.settings.setValue("only_video", config.params['only_video'])
         self.settings.setValue("tts_type", config.params['tts_type'])
         self.settings.setValue("clone_api", config.params['clone_api'])
+        self.settings.setValue("chattts_api", config.params['chattts_api'])
         self.settings.setValue("video_autorate", config.params['video_autorate'])
         self.settings.setValue("append_video", config.params['append_video'])
         self.settings.setValue("clone_voicelist", ','.join(config.clone_voicelist))

videotrans/mainwin/subform.py

@@ -271,6 +271,59 @@ def save():
         self.main.clonw.test.clicked.connect(test)
         self.main.clonw.show()
 
+    def set_chattts_address(self):
+        class TestTTS(QThread):
+            uito = Signal(str)
+
+            def __init__(self, *, parent=None, text=None):
+                super().__init__(parent=parent)
+                self.text = text
+
+            def run(self):
+                from videotrans.tts.chattts import get_voice
+                try:
+                    get_voice(text=self.text, role="boy1", set_p=False, filename=config.homedir + "/test.mp3")
+
+                    self.uito.emit("ok")
+                except Exception as e:
+                    self.uito.emit(str(e))
+
+        def feed(d):
+            if d == "ok":
+                tools.pygameaudio(config.homedir + "/test.mp3")
+                QtWidgets.QMessageBox.information(self.main.chatttsw, "ok", "Test Ok")
+            else:
+                QtWidgets.QMessageBox.critical(self.main.chatttsw, config.transobj['anerror'], d)
+            self.main.chatttsw.test.setText('测试' if config.defaulelang == 'zh' else 'Test')
+
+        def test():
+            if not self.main.chatttsw.chattts_address.text().strip():
+                QtWidgets.QMessageBox.critical(self.main.chatttsw, config.transobj['anerror'], '必须填写http地址')
+                return
+            config.params['chattts_api'] = self.main.chatttsw.chattts_address.text().strip()
+            task = TestTTS(parent=self.main.chatttsw,
+                           text="你好啊我的朋友" if config.defaulelang == 'zh' else 'hello,my friend'
+                           )
+            self.main.chatttsw.test.setText('测试中请稍等...' if config.defaulelang == 'zh' else 'Testing...')
+            task.uito.connect(feed)
+            task.start()
+
+        def save():
+            key = self.main.chatttsw.chattts_address.text().strip()
+            key = key.rstrip('/')
+            key = 'http://' + key.replace('http://', '').replace('/tts','')
+            self.main.settings.setValue("chattts_api", key)
+            config.params["chattts_api"] = key
+            self.main.chatttsw.close()
+
+        from videotrans.component import ChatttsForm
+        self.main.chatttsw = ChatttsForm()
+        if config.params["chattts_api"]:
+            self.main.chatttsw.chattts_address.setText(config.params["chattts_api"])
+        self.main.chatttsw.set_chattts.clicked.connect(save)
+        self.main.chatttsw.test.clicked.connect(test)
+        self.main.chatttsw.show()
+
     def set_zh_recogn(self):
         class Test(QThread):
             uito = Signal(str)

videotrans/tts/__init__.py

@@ -43,6 +43,8 @@ def text_to_speech(
         # lasterror=get_voice(text=text, role=role, rate=rate,language=language, filename=filename,set_p=set_p,inst=inst)
     elif tts_type =='gtts':
         from .gtts import get_voice
+    elif tts_type=='ChatTTS':
+        from .chattts import get_voice
         # lasterror=get_voice(text=text, role=role, rate=rate, language=language,filename=filename,set_p=set_p,inst=inst)
     if get_voice:
         get_voice(