remove duplicate spaces

GPT_SoVITS/TTS_infer_pack/TextPreprocessor.py

@@ -121,70 +121,70 @@ class TextPreprocessor:
 
     def get_phones_and_bert(self, text: str, language: str, version: str, final: bool = False):
         with self.bert_lock:
-            if language in {"all_zh", "all_yue", "all_ja", "all_ko", "zh", "ja", "ko", "yue", "en", "auto", "auto_yue"}:
+            text = re.sub(r' {2,}', ' ', text)
-                textlist = []
+            textlist = []
-                langlist = []
+            langlist = []
-                if language == "all_zh":
+            if language == "all_zh":
-                    for tmp in LangSegmenter.getTexts(text,"zh"):
+                for tmp in LangSegmenter.getTexts(text,"zh"):
+                    langlist.append(tmp["lang"])
+                    textlist.append(tmp["text"])
+            elif language == "all_yue":
+                for tmp in LangSegmenter.getTexts(text,"zh"):
+                    if tmp["lang"] == "zh":
+                        tmp["lang"] = "yue"
+                    langlist.append(tmp["lang"])
+                    textlist.append(tmp["text"])
+            elif language == "all_ja":
+                for tmp in LangSegmenter.getTexts(text,"ja"):
+                    langlist.append(tmp["lang"])
+                    textlist.append(tmp["text"])
+            elif language == "all_ko":
+                for tmp in LangSegmenter.getTexts(text,"ko"):
+                    langlist.append(tmp["lang"])
+                    textlist.append(tmp["text"])
+            elif language == "en":
+                formattext = text
+                while "  " in formattext:
+                    formattext = formattext.replace("  ", " ")
+                langlist.append("en")
+                textlist.append(formattext)
+            elif language == "auto":
+                for tmp in LangSegmenter.getTexts(text):
+                    langlist.append(tmp["lang"])
+                    textlist.append(tmp["text"])
+            elif language == "auto_yue":
+                for tmp in LangSegmenter.getTexts(text):
+                    if tmp["lang"] == "zh":
+                        tmp["lang"] = "yue"
+                    langlist.append(tmp["lang"])
+                    textlist.append(tmp["text"])
+            else:
+                for tmp in LangSegmenter.getTexts(text):
+                    if langlist:
+                        if (tmp["lang"] == "en" and langlist[-1] == "en") or (tmp["lang"] != "en" and langlist[-1] != "en"):
+                            textlist[-1] += tmp["text"]
+                            continue
+                    if tmp["lang"] == "en":
                         langlist.append(tmp["lang"])
-                        textlist.append(tmp["text"])
+                    else:
-                elif language == "all_yue":
+                        # 因无法区别中日韩文汉字,以用户输入为准
-                    for tmp in LangSegmenter.getTexts(text,"zh"):
+                        langlist.append(language)
-                        if tmp["lang"] == "zh":
+                    textlist.append(tmp["text"])
-                            tmp["lang"] = "yue"
+            # print(textlist)
-                        langlist.append(tmp["lang"])
+            # print(langlist)
-                        textlist.append(tmp["text"])
+            phones_list = []
-                elif language == "all_ja":
+            bert_list = []
-                    for tmp in LangSegmenter.getTexts(text,"ja"):
+            norm_text_list = []
-                        langlist.append(tmp["lang"])
+            for i in range(len(textlist)):
-                        textlist.append(tmp["text"])
+                lang = langlist[i]
-                elif language == "all_ko":
+                phones, word2ph, norm_text = self.clean_text_inf(textlist[i], lang, version)
-                    for tmp in LangSegmenter.getTexts(text,"ko"):
+                bert = self.get_bert_inf(phones, word2ph, norm_text, lang)
-                        langlist.append(tmp["lang"])
+                phones_list.append(phones)
-                        textlist.append(tmp["text"])
+                norm_text_list.append(norm_text)
-                elif language == "en":
+                bert_list.append(bert)
-                    formattext = text
+            bert = torch.cat(bert_list, dim=1)
-                    while "  " in formattext:
+            phones = sum(phones_list, [])
-                        formattext = formattext.replace("  ", " ")
+            norm_text = "".join(norm_text_list)
-                    langlist.append("en")
-                    textlist.append(formattext)
-                elif language == "auto":
-                    for tmp in LangSegmenter.getTexts(text):
-                        langlist.append(tmp["lang"])
-                        textlist.append(tmp["text"])
-                elif language == "auto_yue":
-                    for tmp in LangSegmenter.getTexts(text):
-                        if tmp["lang"] == "zh":
-                            tmp["lang"] = "yue"
-                        langlist.append(tmp["lang"])
-                        textlist.append(tmp["text"])
-                else:
-                    for tmp in LangSegmenter.getTexts(text):
-                        if langlist:
-                            if (tmp["lang"] == "en" and langlist[-1] == "en") or (tmp["lang"] != "en" and langlist[-1] != "en"):
-                                textlist[-1] += tmp["text"]
-                                continue
-                        if tmp["lang"] == "en":
-                            langlist.append(tmp["lang"])
-                        else:
-                            # 因无法区别中日韩文汉字,以用户输入为准
-                            langlist.append(language)
-                        textlist.append(tmp["text"])
-                # print(textlist)
-                # print(langlist)
-                phones_list = []
-                bert_list = []
-                norm_text_list = []
-                for i in range(len(textlist)):
-                    lang = langlist[i]
-                    phones, word2ph, norm_text = self.clean_text_inf(textlist[i], lang, version)
-                    bert = self.get_bert_inf(phones, word2ph, norm_text, lang)
-                    phones_list.append(phones)
-                    norm_text_list.append(norm_text)
-                    bert_list.append(bert)
-                bert = torch.cat(bert_list, dim=1)
-                phones = sum(phones_list, [])
-                norm_text = "".join(norm_text_list)
 
             if not final and len(phones) < 6:
                 return self.get_phones_and_bert("." + text, language, version, final=True)

GPT_SoVITS/inference_webui.py

@@ -586,70 +586,70 @@ from text import chinese
 
 
 def get_phones_and_bert(text, language, version, final=False):
-    if language in {"all_zh", "all_yue", "all_ja", "all_ko", "zh", "ja", "ko", "yue", "en", "auto", "auto_yue"}:
+    text = re.sub(r' {2,}', ' ', text)
-        textlist = []
+    textlist = []
-        langlist = []
+    langlist = []
-        if language == "all_zh":
+    if language == "all_zh":
-            for tmp in LangSegmenter.getTexts(text,"zh"):
+        for tmp in LangSegmenter.getTexts(text,"zh"):
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "all_yue":
+        for tmp in LangSegmenter.getTexts(text,"zh"):
+            if tmp["lang"] == "zh":
+                tmp["lang"] = "yue"
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "all_ja":
+        for tmp in LangSegmenter.getTexts(text,"ja"):
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "all_ko":
+        for tmp in LangSegmenter.getTexts(text,"ko"):
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "en":
+        formattext = text
+        while "  " in formattext:
+            formattext = formattext.replace("  ", " ")
+        langlist.append("en")
+        textlist.append(formattext)
+    elif language == "auto":
+        for tmp in LangSegmenter.getTexts(text):
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "auto_yue":
+        for tmp in LangSegmenter.getTexts(text):
+            if tmp["lang"] == "zh":
+                tmp["lang"] = "yue"
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    else:
+        for tmp in LangSegmenter.getTexts(text):
+            if langlist:
+                if (tmp["lang"] == "en" and langlist[-1] == "en") or (tmp["lang"] != "en" and langlist[-1] != "en"):
+                    textlist[-1] += tmp["text"]
+                    continue
+            if tmp["lang"] == "en":
                 langlist.append(tmp["lang"])
-                textlist.append(tmp["text"])
+            else:
-        elif language == "all_yue":
+                # 因无法区别中日韩文汉字,以用户输入为准
-            for tmp in LangSegmenter.getTexts(text,"zh"):
+                langlist.append(language)
-                if tmp["lang"] == "zh":
+            textlist.append(tmp["text"])
-                    tmp["lang"] = "yue"
+    print(textlist)
-                langlist.append(tmp["lang"])
+    print(langlist)
-                textlist.append(tmp["text"])
+    phones_list = []
-        elif language == "all_ja":
+    bert_list = []
-            for tmp in LangSegmenter.getTexts(text,"ja"):
+    norm_text_list = []
-                langlist.append(tmp["lang"])
+    for i in range(len(textlist)):
-                textlist.append(tmp["text"])
+        lang = langlist[i]
-        elif language == "all_ko":
+        phones, word2ph, norm_text = clean_text_inf(textlist[i], lang, version)
-            for tmp in LangSegmenter.getTexts(text,"ko"):
+        bert = get_bert_inf(phones, word2ph, norm_text, lang)
-                langlist.append(tmp["lang"])
+        phones_list.append(phones)
-                textlist.append(tmp["text"])
+        norm_text_list.append(norm_text)
-        elif language == "en":
+        bert_list.append(bert)
-            formattext = text
+    bert = torch.cat(bert_list, dim=1)
-            while "  " in formattext:
+    phones = sum(phones_list, [])
-                formattext = formattext.replace("  ", " ")
+    norm_text = "".join(norm_text_list)
-            langlist.append("en")
-            textlist.append(formattext)
-        elif language == "auto":
-            for tmp in LangSegmenter.getTexts(text):
-                langlist.append(tmp["lang"])
-                textlist.append(tmp["text"])
-        elif language == "auto_yue":
-            for tmp in LangSegmenter.getTexts(text):
-                if tmp["lang"] == "zh":
-                    tmp["lang"] = "yue"
-                langlist.append(tmp["lang"])
-                textlist.append(tmp["text"])
-        else:
-            for tmp in LangSegmenter.getTexts(text):
-                if langlist:
-                    if (tmp["lang"] == "en" and langlist[-1] == "en") or (tmp["lang"] != "en" and langlist[-1] != "en"):
-                        textlist[-1] += tmp["text"]
-                        continue
-                if tmp["lang"] == "en":
-                    langlist.append(tmp["lang"])
-                else:
-                    # 因无法区别中日韩文汉字,以用户输入为准
-                    langlist.append(language)
-                textlist.append(tmp["text"])
-        print(textlist)
-        print(langlist)
-        phones_list = []
-        bert_list = []
-        norm_text_list = []
-        for i in range(len(textlist)):
-            lang = langlist[i]
-            phones, word2ph, norm_text = clean_text_inf(textlist[i], lang, version)
-            bert = get_bert_inf(phones, word2ph, norm_text, lang)
-            phones_list.append(phones)
-            norm_text_list.append(norm_text)
-            bert_list.append(bert)
-        bert = torch.cat(bert_list, dim=1)
-        phones = sum(phones_list, [])
-        norm_text = "".join(norm_text_list)
 
     if not final and len(phones) < 6:
         return get_phones_and_bert("." + text, language, version, final=True)

api.py

@@ -532,68 +532,68 @@ from text import chinese
 
 
 def get_phones_and_bert(text, language, version, final=False):
-    if language in {"all_zh", "all_yue", "all_ja", "all_ko", "zh", "ja", "ko", "yue", "en", "auto", "auto_yue"}:
+    text = re.sub(r' {2,}', ' ', text)
-        textlist = []
+    textlist = []
-        langlist = []
+    langlist = []
-        if language == "all_zh":
+    if language == "all_zh":
-            for tmp in LangSegmenter.getTexts(text,"zh"):
+        for tmp in LangSegmenter.getTexts(text,"zh"):
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "all_yue":
+        for tmp in LangSegmenter.getTexts(text,"zh"):
+            if tmp["lang"] == "zh":
+                tmp["lang"] = "yue"
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "all_ja":
+        for tmp in LangSegmenter.getTexts(text,"ja"):
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "all_ko":
+        for tmp in LangSegmenter.getTexts(text,"ko"):
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "en":
+        formattext = text
+        while "  " in formattext:
+            formattext = formattext.replace("  ", " ")
+        langlist.append("en")
+        textlist.append(formattext)
+    elif language == "auto":
+        for tmp in LangSegmenter.getTexts(text):
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    elif language == "auto_yue":
+        for tmp in LangSegmenter.getTexts(text):
+            if tmp["lang"] == "zh":
+                tmp["lang"] = "yue"
+            langlist.append(tmp["lang"])
+            textlist.append(tmp["text"])
+    else:
+        for tmp in LangSegmenter.getTexts(text):
+            if langlist:
+                if (tmp["lang"] == "en" and langlist[-1] == "en") or (tmp["lang"] != "en" and langlist[-1] != "en"):
+                    textlist[-1] += tmp["text"]
+                    continue
+            if tmp["lang"] == "en":
                 langlist.append(tmp["lang"])
-                textlist.append(tmp["text"])
+            else:
-        elif language == "all_yue":
+                # 因无法区别中日韩文汉字,以用户输入为准
-            for tmp in LangSegmenter.getTexts(text,"zh"):
+                langlist.append(language)
-                if tmp["lang"] == "zh":
+            textlist.append(tmp["text"])
-                    tmp["lang"] = "yue"
+    phones_list = []
-                langlist.append(tmp["lang"])
+    bert_list = []
-                textlist.append(tmp["text"])
+    norm_text_list = []
-        elif language == "all_ja":
+    for i in range(len(textlist)):
-            for tmp in LangSegmenter.getTexts(text,"ja"):
+        lang = langlist[i]
-                langlist.append(tmp["lang"])
+        phones, word2ph, norm_text = clean_text_inf(textlist[i], lang, version)
-                textlist.append(tmp["text"])
+        bert = get_bert_inf(phones, word2ph, norm_text, lang)
-        elif language == "all_ko":
+        phones_list.append(phones)
-            for tmp in LangSegmenter.getTexts(text,"ko"):
+        norm_text_list.append(norm_text)
-                langlist.append(tmp["lang"])
+        bert_list.append(bert)
-                textlist.append(tmp["text"])
+    bert = torch.cat(bert_list, dim=1)
-        elif language == "en":
+    phones = sum(phones_list, [])
-            formattext = text
+    norm_text = "".join(norm_text_list)
-            while "  " in formattext:
-                formattext = formattext.replace("  ", " ")
-            langlist.append("en")
-            textlist.append(formattext)
-        elif language == "auto":
-            for tmp in LangSegmenter.getTexts(text):
-                langlist.append(tmp["lang"])
-                textlist.append(tmp["text"])
-        elif language == "auto_yue":
-            for tmp in LangSegmenter.getTexts(text):
-                if tmp["lang"] == "zh":
-                    tmp["lang"] = "yue"
-                langlist.append(tmp["lang"])
-                textlist.append(tmp["text"])
-        else:
-            for tmp in LangSegmenter.getTexts(text):
-                if langlist:
-                    if (tmp["lang"] == "en" and langlist[-1] == "en") or (tmp["lang"] != "en" and langlist[-1] != "en"):
-                        textlist[-1] += tmp["text"]
-                        continue
-                if tmp["lang"] == "en":
-                    langlist.append(tmp["lang"])
-                else:
-                    # 因无法区别中日韩文汉字,以用户输入为准
-                    langlist.append(language)
-                textlist.append(tmp["text"])
-        phones_list = []
-        bert_list = []
-        norm_text_list = []
-        for i in range(len(textlist)):
-            lang = langlist[i]
-            phones, word2ph, norm_text = clean_text_inf(textlist[i], lang, version)
-            bert = get_bert_inf(phones, word2ph, norm_text, lang)
-            phones_list.append(phones)
-            norm_text_list.append(norm_text)
-            bert_list.append(bert)
-        bert = torch.cat(bert_list, dim=1)
-        phones = sum(phones_list, [])
-        norm_text = "".join(norm_text_list)
 
     if not final and len(phones) < 6:
         return get_phones_and_bert("." + text, language, version, final=True)