Update infer-web.py

infer-web.py

@@ -58,7 +58,7 @@ torch.manual_seed(114514)
 config = Config()
 i18n = I18nAuto()
 i18n.print()
-# 判断是否有能用来训练和加速推理的N卡
+# Determine whether there is an N card that can be used for training and accelerated reasoning
 ngpu = torch.cuda.device_count()
 gpu_infos = []
 mem = []
@@ -91,7 +91,7 @@ if torch.cuda.is_available() or ngpu != 0:
             ]
         ):
             # A10#A100#V100#A40#P40#M40#K80#A4500
-            if_gpu_ok = True  # 至少有一张能用的N卡
+            if_gpu_ok = True  # At least one usable Nvidia card
             gpu_infos.append("%s\t%s" % (i, gpu_name))
             mem.append(
                 int(
@@ -106,7 +106,7 @@ if if_gpu_ok and len(gpu_infos) > 0:
     gpu_info = "\n".join(gpu_infos)
     default_batch_size = min(mem) // 2
 else:
-    gpu_info = i18n("很遗憾您这没有能用的显卡来支持您训练")
+    gpu_info = i18n("Unfortunately you don't have a working graphics card to support your training.")
     default_batch_size = 1
 gpus = "-".join([i[0] for i in gpu_infos])
 
@@ -197,7 +197,7 @@ def vc_single(
             )
             if file_index != ""
             else file_index2
-        )  # 防止小白写错，自动帮他替换掉
+        )  # Prevent newbies from making mistakes and automatically replace them for them
         # file_big_npy = (
         #     file_big_npy.strip(" ").strip('"').strip("\n").strip('"').strip(" ")
         # )
@@ -263,7 +263,7 @@ def vc_multi(
     try:
         dir_path = (
             dir_path.strip(" ").strip('"').strip("\n").strip('"').strip(" ")
-        )  # 防止小白拷路径头尾带了空格和"和回车
+        )  # Prevent newbies from copying the path with spaces, " and carriage returns at the beginning and end
         opt_root = opt_root.strip(" ").strip('"').strip("\n").strip('"').strip(" ")
         os.makedirs(opt_root, exist_ok=True)
         try:
@@ -402,18 +402,18 @@ def uvr(model_name, inp_root, save_root_vocal, paths, save_root_ins, agg, format
     yield "\n".join(infos)
 
 
-# 一个选项卡全局只能有一个音色
+# A tab can only have one sound globally.
 def get_vc(sid, to_return_protect0, to_return_protect1):
     global n_spk, tgt_sr, net_g, vc, cpt, version
     if sid == "" or sid == []:
         global hubert_model
-        if hubert_model is not None:  # 考虑到轮询, 需要加个判断看是否 sid 是由有模型切换到无模型的
+        if hubert_model is not None:  # Considering polling, we need to add a check to see if the sid is switched from a model to a model-free state.
             print("clean_empty_cache")
             del net_g, n_spk, vc, hubert_model, tgt_sr  # ,cpt
             hubert_model = net_g = n_spk = vc = hubert_model = tgt_sr = None
             if torch.cuda.is_available():
                 torch.cuda.empty_cache()
-            ###楼下不这么折腾清理不干净
+            ###The downstairs won't be clean without this much trouble
             if_f0 = cpt.get("f0", 1)
             version = cpt.get("version", "v1")
             if version == "v1":
@@ -523,8 +523,8 @@ def if_done(done, p):
 
 def if_done_multi(done, ps):
     while 1:
-        # poll==None代表进程未结束
-        # 只要有一个进程未结束都不停
+        # poll==None means the process is not finished
+        # As long as there is a process that has not ended, it will not stop.
         flag = 1
         for p in ps:
             if p.poll() is None:
@@ -549,7 +549,7 @@ def preprocess_dataset(trainset_dir, exp_dir, sr, n_p):
     )
     print(cmd)
     p = Popen(cmd, shell=True)  # , stdin=PIPE, stdout=PIPE,stderr=PIPE,cwd=now_dir
-    ###煞笔gr, popen read都非得全跑完了再一次性读取, 不用gr就正常读一句输出一句;只能额外弄出一个文本流定时读
+    ###Damn, gr and popen read both have to run completely and then read at once. Without gr, it can just read and output one sentence at a time. I can only create an additional text stream to read at a fixed time.
     done = [False]
     threading.Thread(
         target=if_done,
@@ -586,7 +586,7 @@ def extract_f0_feature(gpus, n_p, f0method, if_f0, exp_dir, version19, echl):
         )
         print(cmd)
         p = Popen(cmd, shell=True, cwd=now_dir)  # , stdin=PIPE, stdout=PIPE,stderr=PIPE
-        ###煞笔gr, popen read都非得全跑完了再一次性读取, 不用gr就正常读一句输出一句;只能额外弄出一个文本流定时读
+        ###Damn, gr and popen read both have to run completely and then read at once. Without gr, it can just read and output one sentence at a time. I can only create an additional text stream to read at a fixed time.
         done = [False]
         threading.Thread(
             target=if_done,
@@ -607,7 +607,7 @@ def extract_f0_feature(gpus, n_p, f0method, if_f0, exp_dir, version19, echl):
             log = f.read()
         print(log)
         yield log
-    ####对不同part分别开多进程
+    ####Open multiple processes for different parts
     """
     n_part=int(sys.argv[1])
     i_part=int(sys.argv[2])
@@ -636,7 +636,7 @@ def extract_f0_feature(gpus, n_p, f0method, if_f0, exp_dir, version19, echl):
             cmd, shell=True, cwd=now_dir
         )  # , shell=True, stdin=PIPE, stdout=PIPE, stderr=PIPE, cwd=now_dir
         ps.append(p)
-    ###煞笔gr, popen read都非得全跑完了再一次性读取, 不用gr就正常读一句输出一句;只能额外弄出一个文本流定时读
+    ###Damn, gr and popen read both have to run completely and then read at once. Without gr, it can just read and output one sentence at a time. I can only create an additional text stream to read at a fixed time.
     done = [False]
     threading.Thread(
         target=if_done_multi,
@@ -779,7 +779,7 @@ def click_train(
     if_save_every_weights18,
     version19,
 ):
-    # 生成filelist
+    # Generate filelist
     exp_dir = "%s/logs/%s" % (now_dir, exp_dir1)
     os.makedirs(exp_dir, exist_ok=True)
     gt_wavs_dir = "%s/0_gt_wavs" % (exp_dir)
@@ -846,7 +846,7 @@ def click_train(
     with open("%s/filelist.txt" % exp_dir, "w") as f:
         f.write("\n".join(opt))
     print("write filelist done")
-    # 生成config#无需生成config
+    # Generate config#No need to generate config
     # cmd = python_cmd + " train_nsf_sim_cache_sid_load_pretrain.py -e mi-test -sr 40k -f0 1 -bs 4 -g 0 -te 10 -se 5 -pg pretrained/f0G40k.pth -pd pretrained/f0D40k.pth -l 1 -c 0"
     print("use gpus:", gpus16)
     if pretrained_G14 == "":
@@ -867,9 +867,9 @@ def click_train(
                 save_epoch10,
                 "-pg %s" % pretrained_G14 if pretrained_G14 != "" else "",
                 "-pd %s" % pretrained_D15 if pretrained_D15 != "" else "",
-                1 if if_save_latest13 == i18n("是") else 0,
-                1 if if_cache_gpu17 == i18n("是") else 0,
-                1 if if_save_every_weights18 == i18n("是") else 0,
+                1 if if_save_latest13 == i18n("yes") else 0,
+                1 if if_cache_gpu17 == i18n("yes") else 0,
+                1 if if_save_every_weights18 == i18n("yes") else 0,
                 version19,
             )
         )
@@ -886,16 +886,16 @@ def click_train(
                 save_epoch10,
                 "-pg %s" % pretrained_G14 if pretrained_G14 != "" else "\b",
                 "-pd %s" % pretrained_D15 if pretrained_D15 != "" else "\b",
-                1 if if_save_latest13 == i18n("是") else 0,
-                1 if if_cache_gpu17 == i18n("是") else 0,
-                1 if if_save_every_weights18 == i18n("是") else 0,
+                1 if if_save_latest13 == i18n("yes") else 0,
+                1 if if_cache_gpu17 == i18n("yes") else 0,
+                1 if if_save_every_weights18 == i18n("yes") else 0,
                 version19,
             )
         )
     print(cmd)
     p = Popen(cmd, shell=True, cwd=now_dir)
     p.wait()
-    return "训练结束, 您可查看控制台训练日志或实验文件夹下的train.log"
+    return "After the training is completed, you can view the console training log or train.log in the experiment folder."
 
 
 # but4.click(train_index, [exp_dir1], info3)
@@ -908,10 +908,10 @@ def train_index(exp_dir1, version19):
         else "%s/3_feature768" % (exp_dir)
     )
     if not os.path.exists(feature_dir):
-        return "请先进行特征提取!"
+        return "Please perform feature extraction first!"
     listdir_res = list(os.listdir(feature_dir))
     if len(listdir_res) == 0:
-        return "请先进行特征提取！"
+        return "Please perform feature extraction first!"
     infos = []
     npys = []
     for name in sorted(listdir_res):
@@ -971,11 +971,11 @@ def train_index(exp_dir1, version19):
         % (exp_dir, n_ivf, index_ivf.nprobe, exp_dir1, version19),
     )
     infos.append(
-        "成功构建索引，added_IVF%s_Flat_nprobe_%s_%s_%s.index"
+        "Successfully built index, added_IVF%s_Flat_nprobe_%s_%s_%s.index"
         % (n_ivf, index_ivf.nprobe, exp_dir1, version19)
     )
     # faiss.write_index(index, '%s/added_IVF%s_Flat_FastScan_%s.index'%(exp_dir,n_ivf,version19))
-    # infos.append("成功构建索引，added_IVF%s_Flat_FastScan_%s.index"%(n_ivf,version19))
+    # infos.append("Index built successfully，added_IVF%s_Flat_FastScan_%s.index"%(n_ivf,version19))
     yield "\n".join(infos)
 
 
@@ -1017,7 +1017,7 @@ def train1key(
     )
 
     os.makedirs(model_log_dir, exist_ok=True)
-    #########step1:处理数据
+    #########step1:Process data
     open(preprocess_log_path, "w").close()
     cmd = (
         config.python_cmd
@@ -1025,16 +1025,16 @@ def train1key(
         % (trainset_dir4, sr_dict[sr2], np7, model_log_dir)
         + str(config.noparallel)
     )
-    yield get_info_str(i18n("step1:正在处理数据"))
+    yield get_info_str(i18n("step1:Processing data"))
     yield get_info_str(cmd)
     p = Popen(cmd, shell=True)
     p.wait()
     with open(preprocess_log_path, "r") as f:
         print(f.read())
-    #########step2a:提取音高
+    #########step2a:extract pitch
     open(extract_f0_feature_log_path, "w")
     if if_f0_3:
-        yield get_info_str("step2a:正在提取音高")
+        yield get_info_str("step2a:Extracting pitch")
         cmd = config.python_cmd + " extract_f0_print.py %s %s %s %s" % (
             model_log_dir,
             np7,
@@ -1047,9 +1047,9 @@ def train1key(
         with open(extract_f0_feature_log_path, "r") as f:
             print(f.read())
     else:
-        yield get_info_str(i18n("step2a:无需提取音高"))
-    #######step2b:提取特征
-    yield get_info_str(i18n("step2b:正在提取特征"))
+        yield get_info_str(i18n("step2a:No need to extract pitch"))
+    #######step2b:Extract features
+    yield get_info_str(i18n("step2b:Extracting features"))
     gpus = gpus16.split("-")
     leng = len(gpus)
     ps = []
@@ -1071,9 +1071,9 @@ def train1key(
         p.wait()
     with open(extract_f0_feature_log_path, "r") as f:
         print(f.read())
-    #######step3a:训练模型
-    yield get_info_str(i18n("step3a:正在训练模型"))
-    # 生成filelist
+    #######step3a:Training model
+    yield get_info_str(i18n("step3a:Training model"))
+    # Generate filelist
     if if_f0_3:
         f0_dir = "%s/2a_f0" % model_log_dir
         f0nsf_dir = "%s/2b-f0nsf" % model_log_dir
@@ -1174,8 +1174,8 @@ def train1key(
     yield get_info_str(cmd)
     p = Popen(cmd, shell=True, cwd=now_dir)
     p.wait()
-    yield get_info_str(i18n("训练结束, 您可查看控制台训练日志或实验文件夹下的train.log"))
-    #######step3b:训练索引
+    yield get_info_str(i18n("After the training is completed, you can view the console training log or train.log in the experiment folder."))
+    #######step3b:training index
     npys = []
     listdir_res = list(os.listdir(feature_dir))
     for name in sorted(listdir_res):
@@ -1234,10 +1234,10 @@ def train1key(
         % (model_log_dir, n_ivf, index_ivf.nprobe, exp_dir1, version19),
     )
     yield get_info_str(
-        "成功构建索引, added_IVF%s_Flat_nprobe_%s_%s_%s.index"
+        "Index built successfully, added_IVF%s_Flat_nprobe_%s_%s_%s.index"
         % (n_ivf, index_ivf.nprobe, exp_dir1, version19)
     )
-    yield get_info_str(i18n("全流程结束！"))
+    yield get_info_str(i18n("The whole process is completed!"))
 
 
 #                    ckpt_path2.change(change_info_,[ckpt_path2],[sr__,if_f0__])
@@ -1263,24 +1263,24 @@ def export_onnx(ModelPath, ExportedPath):
     vec_channels = 256 if cpt.get("version", "v1") == "v1" else 768
 
     test_phone = torch.rand(1, 200, vec_channels)  # hidden unit
-    test_phone_lengths = torch.tensor([200]).long()  # hidden unit 长度（貌似没啥用）
-    test_pitch = torch.randint(size=(1, 200), low=5, high=255)  # 基频（单位赫兹）
-    test_pitchf = torch.rand(1, 200)  # nsf基频
-    test_ds = torch.LongTensor([0])  # 说话人ID
-    test_rnd = torch.rand(1, 192, 200)  # 噪声（加入随机因子）
+    test_phone_lengths = torch.tensor([200]).long()  # hidden unit Length (seemingly useless)
+    test_pitch = torch.randint(size=(1, 200), low=5, high=255)  # Fundamental frequency (in Hz)
+    test_pitchf = torch.rand(1, 200)  # NSF base frequency
+    test_ds = torch.LongTensor([0])  # Speaker ID
+    test_rnd = torch.rand(1, 192, 200)  # Noise (adding random factors)
 
-    device = "cpu"  # 导出时设备（不影响使用模型）
+    device = "cpu"  # Device when exporting (does not affect the use of the model)
 
 
     net_g = SynthesizerTrnMsNSFsidM(
         *cpt["config"], is_half=False, version=cpt.get("version", "v1")
-    )  # fp32导出（C++要支持fp16必须手动将内存重新排列所以暂时不用fp16）
+    )  # fp32 export (C++ needs to manually rearrange memory to support fp16, so fp16 is not used for the time being)
     net_g.load_state_dict(cpt["weight"], strict=False)
     input_names = ["phone", "phone_lengths", "pitch", "pitchf", "ds", "rnd"]
     output_names = [
         "audio",
     ]
-    # net_g.construct_spkmixmap(n_speaker) 多角色混合轨道导出
+    # net_g.construct_spkmixmap(n_speaker) export of multiple role mixed tracks
     torch.onnx.export(
         net_g,
         (
@@ -1421,9 +1421,9 @@ def cli_train(com):
     total_epoch = int(com[5]) # 10000
     batch_size = int(com[6])
     gpu_card_slot_numbers = com[7]
-    if_save_latest = i18n("是") if (int(com[8]) == 1) else i18n("否")
-    if_cache_gpu = i18n("是") if (int(com[9]) == 1) else i18n("否")
-    if_save_every_weight = i18n("是") if (int(com[10]) == 1) else i18n("否")
+    if_save_latest = i18n("yes") if (int(com[8]) == 1) else i18n("no")
+    if_cache_gpu = i18n("yes") if (int(com[9]) == 1) else i18n("no")
+    if_save_every_weight = i18n("yes") if (int(com[10]) == 1) else i18n("no")
     version = com[11]
 
     pretrained_base = "pretrained/" if version == "v1" else "pretrained_v2/" 
@@ -1621,11 +1621,11 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
     gr.HTML("<h1> The Mangio-RVC-Fork 💻 </h1>")
     gr.Markdown(
         value=i18n(
-            "本软件以MIT协议开源, 作者不对软件具备任何控制力, 使用软件者、传播软件导出的声音者自负全责. <br>如不认可该条款, 则不能使用或引用软件包内任何代码和文件. 详见根目录<b>使用需遵守的协议-LICENSE.txt</b>."
+            "This software is open source under the MIT agreement. The author does not have any control over the software. Those who use the software and spread the sounds derived from the software are fully responsible. <br>If you do not agree with this clause, you cannot use or reference any code and files in the software package. For details, see the root directory <b>Agreement to be followed-LICENSE.txt</b>."
         )
     )
     with gr.Tabs():
-        with gr.TabItem(i18n("模型推理")):
+        with gr.TabItem(i18n("Model reasoning")):
             # Inference Preset Row
             # with gr.Row():
             #     mangio_preset = gr.Dropdown(label="Inference Preset", choices=sorted(get_presets()))
@@ -1636,14 +1636,14 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
 
             # Other RVC stuff
             with gr.Row():
-                sid0 = gr.Dropdown(label=i18n("推理音色"), choices=sorted(names))
-                refresh_button = gr.Button(i18n("刷新音色列表和索引路径"), variant="primary")
-                clean_button = gr.Button(i18n("卸载音色省显存"), variant="primary")
+                sid0 = gr.Dropdown(label=i18n("Mystery Tone"), choices=sorted(names))
+                refresh_button = gr.Button(i18n("Refresh the patch list and index path"), variant="primary")
+                clean_button = gr.Button(i18n("Uninstalling sounds saves video memory"), variant="primary")
                 spk_item = gr.Slider(
                     minimum=0,
                     maximum=2333,
                     step=1,
-                    label=i18n("请选择说话人id"),
+                    label=i18n("Please select the speaker id"),
                     value=0,
                     visible=False,
                     interactive=True,
@@ -1651,20 +1651,20 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                 clean_button.click(fn=clean, inputs=[], outputs=[sid0])
             with gr.Group():
                 gr.Markdown(
-                    value=i18n("男转女推荐+12key, 女转男推荐-12key, 如果音域爆炸导致音色失真也可以自己调整到合适音域. ")
+                    value=i18n("For male to female, +12key is recommended, for female to male, -12key is recommended. If the sound range explodes and causes timbre distortion, you can adjust it to the appropriate range yourself.")
                 )
                 with gr.Row():
                     with gr.Column():
                         vc_transform0 = gr.Number(
-                            label=i18n("变调(整数, 半音数量, 升八度12降八度-12)"), value=0
+                            label=i18n("Transpose(integer, number of semitones, octave up 12 octave down -12)"), value=0
                         )
                         input_audio0 = gr.Textbox(
-                            label=i18n("输入待处理音频文件路径(默认是正确格式示例)"),
-                            value="E:\\codes\\py39\\test-20230416b\\todo-songs\\冬之花clip1.wav",
+                            label=i18n("Enter the path of the audio file to be processed (the default is an example of the correct format)"),
+                            value="E:\\codes\\py39\\test-20230416b\\todo-songs\\winter flowers clip1.wav",
                         )
                         f0method0 = gr.Radio(
                             label=i18n(
-                                "选择音高提取算法,输入歌声可用pm提速,harvest低音好但巨慢无比,crepe效果好但吃GPU"
+                                "Select the pitch extraction algorithm. You can use pm to speed up the input singing voice. Harvest has good bass but is extremely slow. Crepe has good effect but consumes GPU."
                             ),
                             choices=["pm", "harvest", "dio", "crepe", "crepe-tiny", "mangio-crepe", "mangio-crepe-tiny"], # Fork Feature. Add Crepe-Tiny
                             value="pm",
@@ -1681,19 +1681,19 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         filter_radius0 = gr.Slider(
                             minimum=0,
                             maximum=7,
-                            label=i18n(">=3则使用对harvest音高识别的结果使用中值滤波，数值为滤波半径，使用可以削弱哑音"),
+                            label=i18n(">=3, use median filtering on the result of harvest pitch recognition, the value is the filter radius, which can reduce mute"),
                             value=3,
                             step=1,
                             interactive=True,
                         )
                     with gr.Column():
                         file_index1 = gr.Textbox(
-                            label=i18n("特征检索库文件路径,为空则使用下拉的选择结果"),
+                            label=i18n("Feature retrieval library file path, if empty, use the drop-down selection result"),
                             value="",
                             interactive=True,
                         )
                         file_index2 = gr.Dropdown(
-                            label=i18n("自动检测index路径,下拉式选择(dropdown)"),
+                            label=i18n("Automatically detect index path, drop-down selection"),
                             choices=sorted(index_paths),
                             interactive=True,
                         )
@@ -1701,14 +1701,14 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                             fn=change_choices, inputs=[], outputs=[sid0, file_index2]
                         )
                         # file_big_npy1 = gr.Textbox(
-                        #     label=i18n("特征文件路径"),
+                        #     label=i18n("Feature file path"),
                         #     value="E:\\codes\py39\\vits_vc_gpu_train\\logs\\mi-test-1key\\total_fea.npy",
                         #     interactive=True,
                         # )
                         index_rate1 = gr.Slider(
                             minimum=0,
                             maximum=1,
-                            label=i18n("检索特征占比"),
+                            label=i18n("Search feature proportion"),
                             value=0.88,
                             interactive=True,
                         )
@@ -1716,7 +1716,7 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         resample_sr0 = gr.Slider(
                             minimum=0,
                             maximum=48000,
-                            label=i18n("后处理重采样至最终采样率，0为不进行重采样"),
+                            label=i18n("Post-processing resampling to the final sampling rate, 0 means no resampling"),
                             value=0,
                             step=1,
                             interactive=True,
@@ -1724,7 +1724,7 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         rms_mix_rate0 = gr.Slider(
                             minimum=0,
                             maximum=1,
-                            label=i18n("输入源音量包络替换输出音量包络融合比例，越靠近1越使用输出包络"),
+                            label=i18n("The input source volume envelope replaces the output volume envelope blending ratio. The closer it is to 1, the more the output envelope is used."),
                             value=1,
                             interactive=True,
                         )
@@ -1732,17 +1732,17 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                             minimum=0,
                             maximum=0.5,
                             label=i18n(
-                                "保护清辅音和呼吸声，防止电音撕裂等artifact，拉满0.5不开启，调低加大保护力度但可能降低索引效果"
+                                "Protects clear consonants and breathing sounds, and prevents electronic music tearing and other artifacts. It is not enabled when it is set to 0.5. It is more effective when it is lowered, but the indexing effect may be reduced."
                             ),
                             value=0.33,
                             step=0.01,
                             interactive=True,
                         )
-                    f0_file = gr.File(label=i18n("F0曲线文件, 可选, 一行一个音高, 代替默认F0及升降调"))
-                    but0 = gr.Button(i18n("转换"), variant="primary")
+                    f0_file = gr.File(label=i18n("F0 curve file, optional, one line per pitch, replaces the default F0 and sharp and flat tones"))
+                    but0 = gr.Button(i18n("Convert"), variant="primary")
                     with gr.Row():
-                        vc_output1 = gr.Textbox(label=i18n("输出信息"))
-                        vc_output2 = gr.Audio(label=i18n("输出音频(右下角三个点,点了可以下载)"))
+                        vc_output1 = gr.Textbox(label=i18n("Output information"))
+                        vc_output2 = gr.Audio(label=i18n("Output audio (three dots in the lower right corner, click to download)"))
                     but0.click(
                         vc_single,
                         [
@@ -1765,17 +1765,17 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                     )
             with gr.Group():
                 gr.Markdown(
-                    value=i18n("批量转换, 输入待转换音频文件夹, 或上传多个音频文件, 在指定文件夹(默认opt)下输出转换的音频. ")
+                    value=i18n("Batch conversion, input the audio folder to be converted, or upload multiple audio files, and output the converted audio in the specified folder (default opt).")
                 )
                 with gr.Row():
                     with gr.Column():
                         vc_transform1 = gr.Number(
-                            label=i18n("变调(整数, 半音数量, 升八度12降八度-12)"), value=0
+                            label=i18n("Transpose(integer, number of semitones, octave up 12 octave down -12)"), value=0
                         )
-                        opt_input = gr.Textbox(label=i18n("指定输出文件夹"), value="opt")
+                        opt_input = gr.Textbox(label=i18n("Specify output folder"), value="opt")
                         f0method1 = gr.Radio(
                             label=i18n(
-                                "选择音高提取算法,输入歌声可用pm提速,harvest低音好但巨慢无比,crepe效果好但吃GPU"
+                                "Select the pitch extraction algorithm. You can use pm to speed up the input singing voice. Harvest has good bass but is extremely slow. Crepe has good effect but consumes GPU."
                             ),
                             choices=["pm", "harvest", "crepe"],
                             value="pm",
@@ -1784,19 +1784,19 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         filter_radius1 = gr.Slider(
                             minimum=0,
                             maximum=7,
-                            label=i18n(">=3则使用对harvest音高识别的结果使用中值滤波，数值为滤波半径，使用可以削弱哑音"),
+                            label=i18n(">=3, use median filtering on the result of harvest pitch recognition, the value is the filter radius, which can reduce mute"),
                             value=3,
                             step=1,
                             interactive=True,
                         )
                     with gr.Column():
                         file_index3 = gr.Textbox(
-                            label=i18n("特征检索库文件路径,为空则使用下拉的选择结果"),
+                            label=i18n("Feature retrieval library file path, if empty, use the drop-down selection result"),
                             value="",
                             interactive=True,
                         )
                         file_index4 = gr.Dropdown(
-                            label=i18n("自动检测index路径,下拉式选择(dropdown)"),
+                            label=i18n("Automatically detect index path, drop-down selection"),
                             choices=sorted(index_paths),
                             interactive=True,
                         )
@@ -1806,14 +1806,14 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                             outputs=file_index4,
                         )
                         # file_big_npy2 = gr.Textbox(
-                        #     label=i18n("特征文件路径"),
+                        #     label=i18n("Feature file path"),
                         #     value="E:\\codes\\py39\\vits_vc_gpu_train\\logs\\mi-test-1key\\total_fea.npy",
                         #     interactive=True,
                         # )
                         index_rate2 = gr.Slider(
                             minimum=0,
                             maximum=1,
-                            label=i18n("检索特征占比"),
+                            label=i18n("Search feature proportion"),
                             value=1,
                             interactive=True,
                         )
@@ -1821,7 +1821,7 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         resample_sr1 = gr.Slider(
                             minimum=0,
                             maximum=48000,
-                            label=i18n("后处理重采样至最终采样率，0为不进行重采样"),
+                            label=i18n("Post-processing resampling to the final sampling rate, 0 means no resampling"),
                             value=0,
                             step=1,
                             interactive=True,
@@ -1829,7 +1829,7 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         rms_mix_rate1 = gr.Slider(
                             minimum=0,
                             maximum=1,
-                            label=i18n("输入源音量包络替换输出音量包络融合比例，越靠近1越使用输出包络"),
+                            label=i18n("The input source volume envelope replaces the output volume envelope blending ratio. The closer it is to 1, the more the output envelope is used."),
                             value=1,
                             interactive=True,
                         )
@@ -1837,7 +1837,7 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                             minimum=0,
                             maximum=0.5,
                             label=i18n(
-                                "保护清辅音和呼吸声，防止电音撕裂等artifact，拉满0.5不开启，调低加大保护力度但可能降低索引效果"
+                                "Protects clear consonants and breathing sounds, and prevents electronic music tearing and other artifacts. It is not enabled when it is set to 0.5. It is more effective when it is lowered, but the indexing effect may be reduced."
                             ),
                             value=0.33,
                             step=0.01,
@@ -1845,21 +1845,21 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         )
                     with gr.Column():
                         dir_input = gr.Textbox(
-                            label=i18n("输入待处理音频文件夹路径(去文件管理器地址栏拷就行了)"),
+                            label=i18n("Enter the path of the audio folder to be processed (just copy it from the address bar of the file manager)"),
                             value="E:\codes\py39\\test-20230416b\\todo-songs",
                         )
                         inputs = gr.File(
-                            file_count="multiple", label=i18n("也可批量输入音频文件, 二选一, 优先读文件夹")
+                            file_count="multiple", label=i18n("You can also batch import audio files, choose one of the two, and read the folder first")
                         )
                     with gr.Row():
                         format1 = gr.Radio(
-                            label=i18n("导出文件格式"),
+                            label=i18n("Export file format"),
                             choices=["wav", "flac", "mp3", "m4a"],
                             value="flac",
                             interactive=True,
                         )
-                        but1 = gr.Button(i18n("转换"), variant="primary")
-                        vc_output3 = gr.Textbox(label=i18n("输出信息"))
+                        but1 = gr.Button(i18n("Convert"), variant="primary")
+                        vc_output3 = gr.Textbox(label=i18n("Output information"))
                     but1.click(
                         vc_multi,
                         [
@@ -1891,54 +1891,54 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
             with gr.Group():
                 gr.Markdown(
                     value=i18n(
-                        "人声伴奏分离批量处理， 使用UVR5模型。 <br>"
-                        "合格的文件夹路径格式举例： E:\\codes\\py39\\vits_vc_gpu\\白鹭霜华测试样例(去文件管理器地址栏拷就行了)。 <br>"
-                        "模型分为三类： <br>"
-                        "1、保留人声：不带和声的音频选这个，对主人声保留比HP5更好。内置HP2和HP3两个模型，HP3可能轻微漏伴奏但对主人声保留比HP2稍微好一丁点； <br>"
-                        "2、仅保留主人声：带和声的音频选这个，对主人声可能有削弱。内置HP5一个模型； <br> "
-                        "3、去混响、去延迟模型（by FoxJoy）：<br>"
-                        "  (1)MDX-Net(onnx_dereverb):对于双通道混响是最好的选择，不能去除单通道混响；<br>"
-                        "&emsp;(234)DeEcho:去除延迟效果。Aggressive比Normal去除得更彻底，DeReverb额外去除混响，可去除单声道混响，但是对高频重的板式混响去不干净。<br>"
-                        "去混响/去延迟，附：<br>"
-                        "1、DeEcho-DeReverb模型的耗时是另外2个DeEcho模型的接近2倍；<br>"
-                        "2、MDX-Net-Dereverb模型挺慢的；<br>"
-                        "3、个人推荐的最干净的配置是先MDX-Net再DeEcho-Aggressive。"
+                        "Batch processing of vocal accompaniment separation, using the UVR5 model.<br>"
+                        "An example of a qualified folder path format: E:\\codes\\py39\\vits_vc_gpu\\White Deer Frost Flower Test Sample (just copy it from the address bar of the file manager). <br>"
+                        "Models are divided into three categories: <br>"
+                        "1. Preserve vocals: Choose this for audio without harmony, it preserves the main vocals better than HP5. There are two models built-in, HP2 and HP3. HP3 may slightly miss the accompaniment but preserves the main vocals a little better than HP2; <br>"
+                        "2. Keep only the main voice: Select this for audio with harmony, which may weaken the main voice. Built-in HP5 model; <br>"
+                        "3. De-reverberation and de-delay model (by FoxJoy):"
+                        "   (1) MDX-Net (onnx_dereverb): is the best choice for dual-channel reverberation, and cannot remove single-channel reverberation;"
+                        "&emsp;(234)DeEcho: Removes delay effects. Aggressive removes delay effects more thoroughly than Normal. DeReverb additionally removes reverberation and can remove mono reverberation, but it cannot completely remove high-frequency plate reverberation. <br>"
+                        "De-reverb/de-delay, attached:<br>"
+                        "1. The time consumption of DeEcho-DeReverb model is nearly twice that of the other two DeEcho models;<br>"
+                        "2. The MDX-Net-Dereverb model is quite slow;<br>"
+                        "3. The cleanest configuration I personally recommend is to use MDX-Net first and then DeEcho-Aggressive."
                     )
                 )
                 with gr.Row():
                     with gr.Column():
                         dir_wav_input = gr.Textbox(
-                            label=i18n("输入待处理音频文件夹路径"),
+                            label=i18n("Enter the path of the audio folder to be processed"),
                             value="E:\\codes\\py39\\test-20230416b\\todo-songs\\todo-songs",
                         )
                         wav_inputs = gr.File(
-                            file_count="multiple", label=i18n("也可批量输入音频文件, 二选一, 优先读文件夹")
+                            file_count="multiple", label=i18n("You can also batch import audio files, choose one of the two, and read the folder first")
                         )
                     with gr.Column():
-                        model_choose = gr.Dropdown(label=i18n("模型"), choices=uvr5_names)
+                        model_choose = gr.Dropdown(label=i18n("Model"), choices=uvr5_names)
                         agg = gr.Slider(
                             minimum=0,
                             maximum=20,
                             step=1,
-                            label="人声提取激进程度",
+                            label="Vocal extraction aggressiveness",
                             value=10,
                             interactive=True,
-                            visible=False,  # 先不开放调整
+                            visible=False,  # Not open for adjustment yet
                         )
                         opt_vocal_root = gr.Textbox(
-                            label=i18n("指定输出主人声文件夹"), value="opt"
+                            label=i18n("Specify the output folder for the lead vocals"), value="opt"
                         )
                         opt_ins_root = gr.Textbox(
-                            label=i18n("指定输出非主人声文件夹"), value="opt"
+                            label=i18n("Specify the folder for outputting non-lead vocals"), value="opt"
                         )
                         format0 = gr.Radio(
-                            label=i18n("导出文件格式"),
+                            label=i18n("Export file format"),
                             choices=["wav", "flac", "mp3", "m4a"],
                             value="flac",
                             interactive=True,
                         )
-                    but2 = gr.Button(i18n("转换"), variant="primary")
-                    vc_output4 = gr.Textbox(label=i18n("输出信息"))
+                    but2 = gr.Button(i18n("Convert"), variant="primary")
+                    vc_output4 = gr.Textbox(label=i18n("Output information"))
                     but2.click(
                         uvr,
                         [
@@ -1952,28 +1952,28 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         ],
                         [vc_output4],
                     )
-        with gr.TabItem(i18n("训练")):
+        with gr.TabItem(i18n("train")):
             gr.Markdown(
                 value=i18n(
-                    "step1: 填写实验配置. 实验数据放在logs下, 每个实验一个文件夹, 需手工输入实验名路径, 内含实验配置, 日志, 训练得到的模型文件. "
+                    "Step 1: Fill in the experimental configuration. The experimental data is placed under logs, one folder for each experiment. You need to manually enter the experiment name path, which contains the experimental configuration, logs, and trained model files."
                 )
             )
             with gr.Row():
-                exp_dir1 = gr.Textbox(label=i18n("输入实验名"), value="mi-test")
+                exp_dir1 = gr.Textbox(label=i18n("Enter experiment name"), value="mi-test")
                 sr2 = gr.Radio(
-                    label=i18n("目标采样率"),
+                    label=i18n("target sampling rate"),
                     choices=["40k", "48k"],
                     value="40k",
                     interactive=True,
                 )
                 if_f0_3 = gr.Radio(
-                    label=i18n("模型是否带音高指导(唱歌一定要, 语音可以不要)"),
+                    label=i18n("Does the model have pitch guidance (must be provided for singing, but not for voice)"),
                     choices=[True, False],
                     value=True,
                     interactive=True,
                 )
                 version19 = gr.Radio(
-                    label=i18n("版本"),
+                    label=i18n("Version"),
                     choices=["v1", "v2"],
                     value="v1",
                     interactive=True,
@@ -1983,39 +1983,39 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                     minimum=0,
                     maximum=config.n_cpu,
                     step=1,
-                    label=i18n("提取音高和处理数据使用的CPU进程数"),
+                    label=i18n("The number of CPU processes used to extract pitch and process data"),
                     value=int(np.ceil(config.n_cpu / 1.5)),
                     interactive=True,
                 )
-            with gr.Group():  # 暂时单人的, 后面支持最多4人的#数据处理
+            with gr.Group():  # Currently single-player, will support up to 4 players later#Data processing
                 gr.Markdown(
                     value=i18n(
-                        "step2a: 自动遍历训练文件夹下所有可解码成音频的文件并进行切片归一化, 在实验目录下生成2个wav文件夹; 暂时只支持单人训练. "
+                        "step2a: Automatically traverse all files that can be decoded into audio in the training folder and perform slice normalization, generating 2 wav folders in the experimental directory; currently only supports single-player training."
                     )
                 )
                 with gr.Row():
                     trainset_dir4 = gr.Textbox(
-                        label=i18n("输入训练文件夹路径"), value="E:\\语音音频+标注\\米津玄师\\src"
+                        label=i18n("Enter the training folder path"), value="E:\\Voice Audio+Annotation\\Kenshi Yonezu\\src"
                     )
                     spk_id5 = gr.Slider(
                         minimum=0,
                         maximum=4,
                         step=1,
-                        label=i18n("请指定说话人id"),
+                        label=i18n("Please specify the speaker id"),
                         value=0,
                         interactive=True,
                     )
-                    but1 = gr.Button(i18n("处理数据"), variant="primary")
-                    info1 = gr.Textbox(label=i18n("输出信息"), value="")
+                    but1 = gr.Button(i18n("Process data"), variant="primary")
+                    info1 = gr.Textbox(label=i18n("Output information"), value="")
                     but1.click(
                         preprocess_dataset, [trainset_dir4, exp_dir1, sr2, np7], [info1]
                     )
             with gr.Group():
-                gr.Markdown(value=i18n("step2b: 使用CPU提取音高(如果模型带音高), 使用GPU提取特征(选择卡号)"))
+                gr.Markdown(value=i18n("step2b: Use CPU to extract pitch (if the model has pitch), use GPU to extract features (select card number)"))
                 with gr.Row():
                     with gr.Column():
                         gpus6 = gr.Textbox(
-                            label=i18n("以-分隔输入使用的卡号, 例如   0-1-2   使用卡0和卡1和卡2"),
+                            label=i18n("Enter the card numbers to be used separated by -, for example 0-1-2 uses card 0, card 1 and card 2"),
                             value=gpus,
                             interactive=True,
                         )
@@ -2023,7 +2023,7 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                     with gr.Column():
                         f0method8 = gr.Radio(
                             label=i18n(
-                                "选择音高提取算法:输入歌声可用pm提速,高质量语音但CPU差可用dio提速,harvest质量更好但慢"
+                                "Select the pitch extraction algorithm: input singing voice can be accelerated by pm, high-quality voice but poor CPU can be accelerated by dio, harvest has better quality but is slow"
                             ),
                             choices=["pm", "harvest", "dio", "crepe", "mangio-crepe"], # Fork feature: Crepe on f0 extraction for training.
                             value="harvest",
@@ -2037,21 +2037,21 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                             value=64,
                             interactive=True
                         )
-                    but2 = gr.Button(i18n("特征提取"), variant="primary")
-                    info2 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                    but2 = gr.Button(i18n("Feature extraction"), variant="primary")
+                    info2 = gr.Textbox(label=i18n("Output information"), value="", max_lines=8)
                     but2.click(
                         extract_f0_feature,
                         [gpus6, np7, f0method8, if_f0_3, exp_dir1, version19, extraction_crepe_hop_length],
                         [info2],
                     )
             with gr.Group():
-                gr.Markdown(value=i18n("step3: 填写训练设置, 开始训练模型和索引"))
+                gr.Markdown(value=i18n("step3: Fill in the training settings and start training the model and indexing"))
                 with gr.Row():
                     save_epoch10 = gr.Slider(
                         minimum=0,
                         maximum=50,
                         step=1,
-                        label=i18n("保存频率save_every_epoch"),
+                        label=i18n("Save frequency save_every_epoch"),
                         value=5,
                         interactive=True,
                     )
@@ -2059,7 +2059,7 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         minimum=0,
                         maximum=10000,
                         step=1,
-                        label=i18n("总训练轮数total_epoch"),
+                        label=i18n("Total number of training rounds total_epoch"),
                         value=20,
                         interactive=True,
                     )
@@ -2067,38 +2067,38 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         minimum=1,
                         maximum=40,
                         step=1,
-                        label=i18n("每张显卡的batch_size"),
+                        label=i18n("batch_size for each graphics card"),
                         value=default_batch_size,
                         interactive=True,
                     )
                     if_save_latest13 = gr.Radio(
-                        label=i18n("是否仅保存最新的ckpt文件以节省硬盘空间"),
-                        choices=[i18n("是"), i18n("否")],
-                        value=i18n("否"),
+                        label=i18n("Whether to save only the latest ckpt file to save hard disk space"),
+                        choices=[i18n("yes"), i18n("yes")],
+                        value=i18n("yes"),
                         interactive=True,
                     )
                     if_cache_gpu17 = gr.Radio(
                         label=i18n(
-                            "是否缓存所有训练集至显存. 10min以下小数据可缓存以加速训练, 大数据缓存会炸显存也加不了多少速"
+                            "Whether to cache all training sets to the video memory. Small data under 10 minutes can be cached to speed up training. Large data cache will explode the video memory and will not increase the speed much."
                         ),
-                        choices=[i18n("是"), i18n("否")],
-                        value=i18n("否"),
+                        choices=[i18n("yes"), i18n("no")],
+                        value=i18n("no"),
                         interactive=True,
                     )
                     if_save_every_weights18 = gr.Radio(
-                        label=i18n("是否在每次保存时间点将最终小模型保存至weights文件夹"),
-                        choices=[i18n("是"), i18n("否")],
-                        value=i18n("否"),
+                        label=i18n("Whether to save the final small model to the weights folder at each save time point"),
+                        choices=[i18n("yes"), i18n("no")],
+                        value=i18n("no"),
                         interactive=True,
                     )
                 with gr.Row():
                     pretrained_G14 = gr.Textbox(
-                        label=i18n("加载预训练底模G路径"),
+                        label=i18n("Load the pre-trained bottom model G path"),
                         value="pretrained/f0G40k.pth",
                         interactive=True,
                     )
                     pretrained_D15 = gr.Textbox(
-                        label=i18n("加载预训练底模D路径"),
+                        label=i18n("Load the pre-trained bottom model D path"),
                         value="pretrained/f0D40k.pth",
                         interactive=True,
                     )
@@ -2118,14 +2118,14 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         [f0method8, pretrained_G14, pretrained_D15],
                     )
                     gpus16 = gr.Textbox(
-                        label=i18n("以-分隔输入使用的卡号, 例如   0-1-2   使用卡0和卡1和卡2"),
+                        label=i18n("Enter the card numbers to be used separated by -, for example 0-1-2 uses card 0, card 1 and card 2"),
                         value=gpus,
                         interactive=True,
                     )
-                    but3 = gr.Button(i18n("训练模型"), variant="primary")
-                    but4 = gr.Button(i18n("训练特征索引"), variant="primary")
-                    but5 = gr.Button(i18n("一键训练"), variant="primary")
-                    info3 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=10)
+                    but3 = gr.Button(i18n("Training model"), variant="primary")
+                    but4 = gr.Button(i18n("Training feature index"), variant="primary")
+                    but5 = gr.Button(i18n("One click training"), variant="primary")
+                    info3 = gr.Textbox(label=i18n("Output information"), value="", max_lines=10)
                     but3.click(
                         click_train,
                         [
@@ -2172,50 +2172,50 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                         info3,
                     )
 
-        with gr.TabItem(i18n("ckpt处理")):
+        with gr.TabItem(i18n("ckpt processing")):
             with gr.Group():
-                gr.Markdown(value=i18n("模型融合, 可用于测试音色融合"))
+                gr.Markdown(value=i18n("Model fusion, can be used to test timbre fusion"))
                 with gr.Row():
-                    ckpt_a = gr.Textbox(label=i18n("A模型路径"), value="", interactive=True)
-                    ckpt_b = gr.Textbox(label=i18n("B模型路径"), value="", interactive=True)
+                    ckpt_a = gr.Textbox(label=i18n("A model path"), value="", interactive=True)
+                    ckpt_b = gr.Textbox(label=i18n("B model path"), value="", interactive=True)
                     alpha_a = gr.Slider(
                         minimum=0,
                         maximum=1,
-                        label=i18n("A模型权重"),
+                        label=i18n("A model weight"),
                         value=0.5,
                         interactive=True,
                     )
                 with gr.Row():
                     sr_ = gr.Radio(
-                        label=i18n("目标采样率"),
+                        label=i18n("target sampling rate"),
                         choices=["40k", "48k"],
                         value="40k",
                         interactive=True,
                     )
                     if_f0_ = gr.Radio(
-                        label=i18n("模型是否带音高指导"),
-                        choices=[i18n("是"), i18n("否")],
-                        value=i18n("是"),
+                        label=i18n("Does the model have pitch guidance?"),
+                        choices=[i18n("yes"), i18n("no")],
+                        value=i18n("yes"),
                         interactive=True,
                     )
                     info__ = gr.Textbox(
-                        label=i18n("要置入的模型信息"), value="", max_lines=8, interactive=True
+                        label=i18n("Model information to be placed"), value="", max_lines=8, interactive=True
                     )
                     name_to_save0 = gr.Textbox(
-                        label=i18n("保存的模型名不带后缀"),
+                        label=i18n("The saved model name has no suffix"),
                         value="",
                         max_lines=1,
                         interactive=True,
                     )
                     version_2 = gr.Radio(
-                        label=i18n("模型版本型号"),
+                        label=i18n("Model version model"),
                         choices=["v1", "v2"],
                         value="v1",
                         interactive=True,
                     )
                 with gr.Row():
-                    but6 = gr.Button(i18n("融合"), variant="primary")
-                    info4 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                    but6 = gr.Button(i18n("Fusion"), variant="primary")
+                    info4 = gr.Textbox(label=i18n("Output information"), value="", max_lines=8)
                 but6.click(
                     merge,
                     [
@@ -2231,71 +2231,71 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                     info4,
                 )  # def merge(path1,path2,alpha1,sr,f0,info):
             with gr.Group():
-                gr.Markdown(value=i18n("修改模型信息(仅支持weights文件夹下提取的小模型文件)"))
+                gr.Markdown(value=i18n("Modify model information (only supports small model files extracted from the weights folder)"))
                 with gr.Row():
                     ckpt_path0 = gr.Textbox(
-                        label=i18n("模型路径"), value="", interactive=True
+                        label=i18n("model path"), value="", interactive=True
                     )
                     info_ = gr.Textbox(
-                        label=i18n("要改的模型信息"), value="", max_lines=8, interactive=True
+                        label=i18n("Model information to be changed"), value="", max_lines=8, interactive=True
                     )
                     name_to_save1 = gr.Textbox(
-                        label=i18n("保存的文件名, 默认空为和源文件同名"),
+                        label=i18n("The saved file name, the default is empty and the same as the source file name"),
                         value="",
                         max_lines=8,
                         interactive=True,
                     )
                 with gr.Row():
-                    but7 = gr.Button(i18n("修改"), variant="primary")
-                    info5 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                    but7 = gr.Button(i18n("Revise"), variant="primary")
+                    info5 = gr.Textbox(label=i18n("Output information"), value="", max_lines=8)
                 but7.click(change_info, [ckpt_path0, info_, name_to_save1], info5)
             with gr.Group():
-                gr.Markdown(value=i18n("查看模型信息(仅支持weights文件夹下提取的小模型文件)"))
+                gr.Markdown(value=i18n("View model information (only supports small model files extracted from the weights folder)"))
                 with gr.Row():
                     ckpt_path1 = gr.Textbox(
-                        label=i18n("模型路径"), value="", interactive=True
+                        label=i18n("model path"), value="", interactive=True
                     )
-                    but8 = gr.Button(i18n("查看"), variant="primary")
-                    info6 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                    but8 = gr.Button(i18n("Check"), variant="primary")
+                    info6 = gr.Textbox(label=i18n("Output information"), value="", max_lines=8)
                 but8.click(show_info, [ckpt_path1], info6)
             with gr.Group():
                 gr.Markdown(
                     value=i18n(
-                        "模型提取(输入logs文件夹下大文件模型路径),适用于训一半不想训了模型没有自动提取保存小文件模型,或者想测试中间模型的情况"
+                        "Model extraction (enter the large file model path in the logs folder), suitable for the case where you don’t want to train the model halfway through and there is no automatic extraction to save the small file model, or you want to test the intermediate model"
                     )
                 )
                 with gr.Row():
                     ckpt_path2 = gr.Textbox(
-                        label=i18n("模型路径"),
+                        label=i18n("model path"),
                         value="E:\\codes\\py39\\logs\\mi-test_f0_48k\\G_23333.pth",
                         interactive=True,
                     )
                     save_name = gr.Textbox(
-                        label=i18n("保存名"), value="", interactive=True
+                        label=i18n("save name"), value="", interactive=True
                     )
                     sr__ = gr.Radio(
-                        label=i18n("目标采样率"),
+                        label=i18n("target sampling rate"),
                         choices=["32k", "40k", "48k"],
                         value="40k",
                         interactive=True,
                     )
                     if_f0__ = gr.Radio(
-                        label=i18n("模型是否带音高指导,1是0否"),
+                        label=i18n("Whether the model has pitch guidance, 1 for yes, 0 for no"),
                         choices=["1", "0"],
                         value="1",
                         interactive=True,
                     )
                     version_1 = gr.Radio(
-                        label=i18n("模型版本型号"),
+                        label=i18n("Model version model"),
                         choices=["v1", "v2"],
                         value="v2",
                         interactive=True,
                     )
                     info___ = gr.Textbox(
-                        label=i18n("要置入的模型信息"), value="", max_lines=8, interactive=True
+                        label=i18n("Model information to be placed"), value="", max_lines=8, interactive=True
                     )
-                    but9 = gr.Button(i18n("提取"), variant="primary")
-                    info7 = gr.Textbox(label=i18n("输出信息"), value="", max_lines=8)
+                    but9 = gr.Button(i18n("extract"), variant="primary")
+                    info7 = gr.Textbox(label=i18n("Output information"), value="", max_lines=8)
                     ckpt_path2.change(
                         change_info_, [ckpt_path2], [sr__, if_f0__, version_1]
                     )
@@ -2305,23 +2305,23 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
                     info7,
                 )
 
-        with gr.TabItem(i18n("Onnx导出")):
+        with gr.TabItem(i18n("Onnx export")):
             with gr.Row():
-                ckpt_dir = gr.Textbox(label=i18n("RVC模型路径"), value="", interactive=True)
+                ckpt_dir = gr.Textbox(label=i18n("RVC model path"), value="", interactive=True)
             with gr.Row():
                 onnx_dir = gr.Textbox(
-                    label=i18n("Onnx输出路径"), value="", interactive=True
+                    label=i18n("Onnx output path"), value="", interactive=True
                 )
             with gr.Row():
                 infoOnnx = gr.Label(label="info")
             with gr.Row():
-                butOnnx = gr.Button(i18n("导出Onnx模型"), variant="primary")
+                butOnnx = gr.Button(i18n("Exporting Onnx Models"), variant="primary")
             butOnnx.click(export_onnx, [ckpt_dir, onnx_dir], infoOnnx)
 
-        tab_faq = i18n("常见问题解答")
+        tab_faq = i18n("FAQ")
         with gr.TabItem(tab_faq):
             try:
-                if tab_faq == "常见问题解答":
+                if tab_faq == "FAQ":
                     with open("docs/faq.md", "r", encoding="utf8") as f:
                         info = f.read()
                 else:
@@ -2453,9 +2453,9 @@ with gr.Blocks(theme=gr.themes.Soft()) as app:
     # )
     #endregion
 
-        # with gr.TabItem(i18n("招募音高曲线前端编辑器")):
-        #     gr.Markdown(value=i18n("加开发群联系我xxxxx"))
-        # with gr.TabItem(i18n("点击查看交流、问题反馈群号")):
+        # with gr.TabItem(i18n("Recruiting pitch curve front-end editor")):
+        #     gr.Markdown(value=i18n("Add the development group to contact me xxxxx"))
+        # with gr.TabItem(i18n("Click to view the communication and problem feedback group number")):
         #     gr.Markdown(value=i18n("xxxxx"))
 
     if config.iscolab or config.paperspace: # Share gradio link for colab and paperspace (FORK FEATURE)