Fix: ui fix

.env

@@ -1,3 +1,4 @@
 HTTP_PROXY=
 WEB_ADDRESS=127.0.0.1:9988
+ENABLE_STS=0
 DEVICE=CUDA

.gitignore

@@ -46,4 +46,5 @@ static/tmp/*.wav
 static/ttslist/*.wav
 *.pth
 *.out
-*.bin
+*.bin
+*.7z

app.py

@@ -23,6 +23,8 @@
 load_dotenv()
 
 web_address = os.getenv('WEB_ADDRESS', '127.0.0.1:9988')
+enable_sts = int(os.getenv('ENABLE_STS', '0'))
+
 
 
 updatecache()
@@ -385,7 +387,7 @@ def onoroff():
     name = request.form.get("name",'')
     status_new = request.form.get("status_new",'')
     if status_new=='on':
-        if not cfg.MYMODEL_OBJS[name] or  isinstance(cfg.MYMODEL_OBJS[name],str):
+        if name not in cfg.MYMODEL_OBJS  or not cfg.MYMODEL_OBJS[name] or  isinstance(cfg.MYMODEL_OBJS[name],str):
             try:
                 print(f'start {name}...')
                 res=logic.load_model(name)
@@ -401,12 +403,17 @@ def onoroff():
         cfg.MYMODEL_OBJS[name]=None
         #删除队列
         cfg.MYMODEL_QUEUE[name]=None
-    return jsonify({"code":0,"msg":"已启动"})
+        return jsonify({"code":0,"msg":"已停止"})
 
 @app.route('/checkupdate', methods=['GET', 'POST'])
 def checkupdate():
     return jsonify({'code': 0, "msg": cfg.updatetips})
 
+@app.route('/stsstatus', methods=['GET', 'POST'])
+def stsstatus():
+    return jsonify({'code': 0, "msg": "start" if cfg.sts_status else "stop"})
+
+
 
 if __name__ == '__main__':
 
@@ -419,24 +426,25 @@ def checkupdate():
         threading.Thread(target=logic.checkupdate).start()
 
         # 如果存在默认模型则启动
+
         if TEXT_MODEL_EXITS:
-            print(langlist['lang2'])
+            print("\n"+langlist['lang2'])
             tts_thread = threading.Thread(target=ttsloop)
             tts_thread.start()
         else:
             app.logger.error(
                 f"\n{langlist['lang3']}: {cfg.download_address}\n")
             input(f"\n{langlist['lang3']}: {cfg.download_address}\n")
             sys.exit()
-
-        if VOICE_MODEL_EXITS:
+        
+        if enable_sts==1 and VOICE_MODEL_EXITS:
             print(langlist['lang4'])
             sts_thread = threading.Thread(target=stsloop)
             sts_thread.start()
-        else:
-            app.logger.error(
-                f"\n{langlist['lang5']}: {cfg.download_address}\n")
-
+        #else:
+        #    app.logger.error(
+        #        f"\n{langlist['lang5']}: {cfg.download_address}\n")
+        
         print(langlist['lang7'])
         try:
             host = web_address.split(':')
@@ -447,4 +455,5 @@ def checkupdate():
     except Exception as e:
         print("error:" + str(e))
         app.logger.error(f"[app]start error:{str(e)}")
+        time.sleep(30)
         sys.exit()

clone/__init__.py

@@ -1,2 +1,2 @@
-VERSION=907
-ver="0.907"
+VERSION=908
+ver="0.908"

clone/cfg.py

@@ -71,6 +71,8 @@ def get_models(path):
 MYMODEL_OBJS=get_models(MYMODEL_DIR)
 MYMODEL_QUEUE={}
 
+sts_status=False
+
 
 device = "cuda" if os.getenv('DEVICE','')=='CUDA' and torch.cuda.is_available() else "cpu"
 q = queue.Queue(maxsize=2000)

clone/logic.py

@@ -50,7 +50,7 @@ def ttsloop():
         if not cfg.setorget_proxy():
             print(f'.env {langlist["lang12"]}')
         else:
-            print(langlist['lang11'])
+            print("\n"+langlist['lang11']+"\n")
         return
     except Exception as e:
         print(f'{langlist["lang13"]}:{str(e)}')
@@ -78,17 +78,21 @@ def ttsloop():
 def stsloop():
     try:
         tts = TTS(model_name='voice_conversion_models/multilingual/vctk/freevc24').to(cfg.device)
-        print(langlist['lang10'])
+        print("\n"+langlist['lang10']+"\n")
     except aiohttp.client_exceptions.ClientOSError as e:
+        cfg.sts_status=False
         print(f'{langlist["lang9"]}：{str(e)}')
         if not cfg.setorget_proxy():
             print(f'.env {langlist["lang12"]}')
         else:
             print(f'{os.environ.get("HTTP_PROXY")} {langlist["lang11"]}')
         return
     except Exception as e:
+        cfg.sts_status=False
         print(f'{langlist["lang9"]}：{str(e)}')
         return
+    else:
+        cfg.sts_status=True
     while 1:
         try:
             obj = cfg.q_sts.get(block=True, timeout=1)        
@@ -196,8 +200,11 @@ def merge_audio_segments(text_list,is_srt=True):
 def openweb(web_address):
     while cfg.tts_n==0:
         time.sleep(5)
-    webbrowser.open("http://"+web_address)
-    print(f"\n{langlist['lang8']} http://{web_address}")
+    try:
+        webbrowser.open("http://"+web_address)
+        print(f"\n{langlist['lang8']} http://{web_address}")
+    except Exception as e:
+        pass
 
 # 判断是否符合字幕格式，如果是，则直接返回
 # 从字幕文件获取格式化后的字幕信息
@@ -346,10 +353,10 @@ def get_subtitle_from_srt0(txt):
 def checkupdate():
     try:
         res=requests.get("https://raw.githubusercontent.com/jianchang512/clone-voice/main/version.json")
-        print(f"{res.status_code=}")
+        #print(f"{res.status_code=}")
         if res.status_code==200:
             d=res.json()
-            print(f"{d=}")
+            #print(f"{d=}")
             if d['version_num']>clone.VERSION:
                 cfg.updatetips=f'New version {d["version"]}'
     except Exception as e:

code_dev.py

@@ -30,7 +30,7 @@ def updatecache():
 
 
 
-ttsv2 = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to(device)
+#ttsv2 = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to(device)
 
 tts = TTS(model_name='voice_conversion_models/multilingual/vctk/freevc24').to(device)
 

runapp.bat

@@ -1,3 +1,5 @@
 @echo off
 
-%cd%\venv\scripts\python.exe %cd%\app.py
+%cd%\venv\scripts\python.exe %cd%\app.py
+
+pause

templates/index.html

@@ -257,6 +257,12 @@ <h4 class="alert-heading d-flex justify-content-between">
             </div>
         </div>
         <div id="sts-area" class="d-none">
+			<div class="p-2">
+				<label for="audioSelect2" class="form-label">
+					<span data-en="The sound file to be clone:" data-cn="要克隆的音色:"></span>
+				</label>
+				<select class="form-select d-inline-block w-auto" id="audioSelect2"></select>
+			</div>
             <div>
                 {% if voice_model %}
                     <div id="dropaudio" class=" border m-2 p-5 text-center "
@@ -275,6 +281,7 @@ <h4 class="alert-heading d-flex justify-content-between">
                             class="d-block" src="/static/images/2.png" width="300">
                     </div>
                 {% endif %}
+				<div class="alert alert-danger d-none" id="sts_no"></div>
             </div>
         </div>
 
@@ -570,7 +577,8 @@ <h4 class="alert-heading d-flex justify-content-between">
                 res.forEach(it => {
                     html += '<option value="' + it + '">' + it + '</option>'
                 })
-                $('#audioSelect').html(html)
+                $('#audioSelect').html(html);
+                $('#audioSelect2').html(html);
             }
         });
     }
@@ -640,6 +648,7 @@ <h4 class="alert-heading d-flex justify-content-between">
                 fun_res(res);
             });
         } else {
+			voice = $("#audioSelect2").val();
             $.post('/sts', {voice: voice, name: window['sts_audio_name']}, function (res) {
                 fun_res(res);
             });
@@ -684,6 +693,25 @@ <h4 class="alert-heading d-flex justify-content-between">
             fileInput.click();
         });
     }
+
+	function sts_status(){
+		$.get('/stsstatus', function (res) {
+            if (res.code === 0) {
+				if(res.msg=='stop'){
+					$('#sts_no').removeClass('d-none').text(language=='zh'?'声音转声音 线程未启动，如果已下载了模型，请打开 .env 文件将 ENABLE_STS=0 改为 ENABLE_STS=1，然后重启软件':'The sts model has not been launched yet, please download it and set .env  ENABLE_STS=0 to ENABLE_STS=1 ');
+					if(!window.sts_time){
+						window.sts_time=0;
+					}
+					window.sts_time+=5
+					if(window.sts_time < 300){
+						setTimeout(sts_status,5000);
+					}
+				}else{
+					$('#sts_no').addClass('d-none').text("")
+				}
+            }
+        });
+	}
 
 
     // 拖拽上传声音音频  
@@ -692,7 +720,7 @@ <h4 class="alert-heading d-flex justify-content-between">
         var formData = new FormData();
         formData.append('audio', file);
         formData.append('save_dir', 'tmp')
-
+		let index=layer.load()
         $.ajax({
             type: 'POST',
             url: '/upload',  // Update with your backend endpoint
@@ -712,9 +740,10 @@ <h4 class="alert-heading d-flex justify-content-between">
                 dropaudio.text('Error');
             },
             beforeSend: function () {
-                dropaudio.text('Upload...');
+                dropaudio.text(language==='zh'?'上传中请稍等..':'Uploading... ');
             },
             complete: function () {
+				layer.close(index)
                 dropaudio.removeClass('dragover');
             }
         });
@@ -777,12 +806,16 @@ <h4 class="alert-heading d-flex justify-content-between">
 			</div>`
         });
     }
+
 
     init();
     stsinit();
     check_start();
     checkupdate();
     switch_language();
+	{% if voice_model %}
+		sts_status();
+	{% endif %}
 
 </script>
 

tts_cache/cache

@@ -1 +1 @@
-{"https://github.com/coqui-ai/TTS/releases/download/v0.13.0_models/speaker_encoder.pt": {"original": "https://github.com/coqui-ai/TTS/releases/download/v0.13.0_models/speaker_encoder.pt", "fn": "d6c67f0e17e8737772c0a7172844fbe40ae535bea18eb69cb212f2adaee668dd", "blocks": true, "time": 1806670143.1189804, "uid": "baefdb5c88dc96918cc04cd7b006b262"}}
+{"https://github.com/coqui-ai/TTS/releases/download/v0.13.0_models/speaker_encoder.pt": {"original": "https://github.com/coqui-ai/TTS/releases/download/v0.13.0_models/speaker_encoder.pt", "fn": "d6c67f0e17e8737772c0a7172844fbe40ae535bea18eb69cb212f2adaee668dd", "blocks": true, "time": 1822501807.932532, "uid": "baefdb5c88dc96918cc04cd7b006b262"}}