汉化

app.py

@@ -29,27 +29,13 @@ def audio_postprocess(self, y):
 
 gr.Audio.postprocess = audio_postprocess
 def create_vc_fn(model, sid):
-    def vc_fn(input_audio, vc_transform, auto_f0, tts_text, tts_voice, tts_mode):
-        if tts_mode:
-            if len(tts_text) > 100 and limitation:
-                return "Text is too long", None
-            if tts_text is None or tts_voice is None:
-                return "You need to enter text and select a voice", None
-            asyncio.run(edge_tts.Communicate(tts_text, "-".join(tts_voice.split('-')[:-1])).save("tts.mp3"))
-            audio, sr = librosa.load("tts.mp3", sr=16000, mono=True)
-            raw_path = io.BytesIO()
-            soundfile.write(raw_path, audio, 16000, format="wav")
-            raw_path.seek(0)
-            out_audio, out_sr = model.infer(sid, vc_transform, raw_path,
-                                            auto_predict_f0=auto_f0,
-                                            )
-            return "Success", (44100, out_audio.cpu().numpy())
+    def vc_fn(input_audio, vc_transform, auto_f0):
         if input_audio is None:
-            return "You need to upload an audio", None
+            return "请先上传音频", None
         sampling_rate, audio = input_audio
         duration = audio.shape[0] / sampling_rate
         if duration > 20 and limitation:
-            return "Please upload an audio file that is less than 20 seconds. If you need to generate a longer audio file, please use Colab.", None
+            return "请上传小于20秒的音频，或点击右上角裁剪", None
         audio = (audio / np.iinfo(audio.dtype).max).astype(np.float32)
         if len(audio.shape) > 1:
             audio = librosa.to_mono(audio.transpose(1, 0))
@@ -64,11 +50,6 @@ def create_vc_fn(model, sid):
         return "Success", (44100, out_audio.cpu().numpy())
     return vc_fn
 
-def change_to_tts_mode(tts_mode):
-    if tts_mode:
-        return gr.Audio.update(visible=False), gr.Textbox.update(visible=True), gr.Dropdown.update(visible=True), gr.Checkbox.update(value=True)
-    else:
-        return gr.Audio.update(visible=True), gr.Textbox.update(visible=False), gr.Dropdown.update(visible=False), gr.Checkbox.update(value=False)
 
 if __name__ == '__main__':
     parser = argparse.ArgumentParser()
@@ -78,12 +59,6 @@ if __name__ == '__main__':
     args = parser.parse_args()
     hubert_model = utils.get_hubert_model().to(args.device)
     models = []
-    # others = {
-    #     "rudolf": "https://huggingface.co/spaces/sayashi/sovits-rudolf",
-    #     "teio": "https://huggingface.co/spaces/sayashi/sovits-teio",
-    #     "goldship": "https://huggingface.co/spaces/sayashi/sovits-goldship",
-    #     "tannhauser": "https://huggingface.co/spaces/sayashi/sovits-tannhauser"
-    # }
     voices = []
     tts_voice_list = asyncio.get_event_loop().run_until_complete(edge_tts.list_voices())
     for r in tts_voice_list:
@@ -95,13 +70,8 @@ if __name__ == '__main__':
         models.append((name, cover, create_vc_fn(model, name)))
     with gr.Blocks() as app:
         gr.Markdown(
-            "# <center> Sovits Models\n"
-            "## <center> The input audio should be clean and pure voice without background music.\n"
-            "![visitor badge](https://visitor-badge.glitch.me/badge?page_id=sayashi.Sovits-Umamusume)\n\n"
-            "[![image](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1wfsBbMzmtLflOJeqc5ZnJiLY7L239hJW?usp=share_link)\n\n"
-            "[![Duplicate this Space](https://huggingface.co/datasets/huggingface/badges/raw/main/duplicate-this-space-sm-dark.svg)](https://huggingface.co/spaces/sayashi/sovits-models?duplicate=true)\n\n"
-            "[![Original Repo](https://badgen.net/badge/icon/github?icon=github&label=Original%20Repo)](https://github.com/svc-develop-team/so-vits-svc)"
-
+            "# <center> GTASA人物SOVITS\n"
+            "## <center> 作者：B站Cyber蝈蝈总\n"
         )
         with gr.Tabs():
             for (name, cover, vc_fn) in models:
@@ -114,28 +84,12 @@ if __name__ == '__main__':
                         )
                     with gr.Row():
                         with gr.Column():
-                            vc_input = gr.Audio(label="Input audio"+' (less than 20 seconds)' if limitation else '')
-                            vc_transform = gr.Number(label="vc_transform", value=0)
-                            auto_f0 = gr.Checkbox(label="auto_f0", value=False)
-                            tts_mode = gr.Checkbox(label="tts (use edge-tts as input)", value=False)
-                            tts_text = gr.Textbox(visible=False, label="TTS text (100 words limitation)" if limitation else "TTS text")
-                            tts_voice = gr.Dropdown(choices=voices, visible=False)
-                            vc_submit = gr.Button("Generate", variant="primary")
+                            vc_input = gr.Audio(label="输入干声"+' (小于 20 秒)' if limitation else '')
+                            vc_transform = gr.Number(label="音高调整(��持正负半音，12为一个八度)", value=0)
+                            auto_f0 = gr.Checkbox(label="自动音高预测（说话模式）", value=False)
+                            vc_submit = gr.Button("生成", variant="primary")
                         with gr.Column():
-                            vc_output1 = gr.Textbox(label="Output Message")
-                            vc_output2 = gr.Audio(label="Output Audio")
-                vc_submit.click(vc_fn, [vc_input, vc_transform, auto_f0, tts_text, tts_voice, tts_mode], [vc_output1, vc_output2])
-                tts_mode.change(change_to_tts_mode, [tts_mode], [vc_input, tts_text, tts_voice, auto_f0])
-            # for category, link in others.items():
-            #     with gr.TabItem(category):
-            #         gr.Markdown(
-            #             f'''
-            #             <center>
-            #               <h2>Click to Go</h2>
-            #               <a href="{link}">
-            #                 <img src="https://huggingface.co/datasets/huggingface/badges/raw/main/open-in-hf-spaces-xl-dark.svg"
-            #               </a>
-            #             </center>
-            #             '''
-            #         )
+                            vc_output1 = gr.Textbox(label="输出信息")
+                            vc_output2 = gr.Audio(label="输出音频")
+                vc_submit.click(vc_fn, [vc_input, vc_transform, auto_f0], [vc_output1, vc_output2])
         app.queue(concurrency_count=1, api_open=args.api).launch(share=args.share)

models/tomori/config_tomori.json

@@ -1,99 +0,0 @@
-{
-  "train": {
-    "log_interval": 200,
-    "eval_interval": 3200,
-    "seed": 1234,
-    "epochs": 30000,
-    "learning_rate": 0.0001,
-    "betas": [
-      0.8,
-      0.99
-    ],
-    "eps": 1e-09,
-    "batch_size": 12,
-    "fp16_run": false,
-    "lr_decay": 0.999875,
-    "segment_size": 10240,
-    "init_lr_ratio": 1,
-    "warmup_epochs": 0,
-    "c_mel": 45,
-    "c_kl": 1.0,
-    "use_sr": true,
-    "max_speclen": 512,
-    "port": "8001",
-    "keep_ckpts": 3,
-    "all_in_mem": false,
-    "vol_aug": true
-  },
-  "data": {
-    "training_files": "filelists/train.txt",
-    "validation_files": "filelists/val.txt",
-    "max_wav_value": 32768.0,
-    "sampling_rate": 44100,
-    "filter_length": 2048,
-    "hop_length": 512,
-    "win_length": 2048,
-    "n_mel_channels": 80,
-    "mel_fmin": 0.0,
-    "mel_fmax": 22050
-  },
-  "model": {
-    "inter_channels": 192,
-    "hidden_channels": 192,
-    "filter_channels": 768,
-    "n_heads": 2,
-    "n_layers": 6,
-    "kernel_size": 3,
-    "p_dropout": 0.1,
-    "resblock": "1",
-    "resblock_kernel_sizes": [
-      3,
-      7,
-      11
-    ],
-    "resblock_dilation_sizes": [
-      [
-        1,
-        3,
-        5
-      ],
-      [
-        1,
-        3,
-        5
-      ],
-      [
-        1,
-        3,
-        5
-      ]
-    ],
-    "upsample_rates": [
-      8,
-      8,
-      2,
-      2,
-      2
-    ],
-    "upsample_initial_channel": 512,
-    "upsample_kernel_sizes": [
-      16,
-      16,
-      4,
-      4,
-      4
-    ],
-    "n_layers_q": 3,
-    "use_spectral_norm": false,
-    "gin_channels": 768,
-    "ssl_dim": 768,
-    "n_speakers": 1,
-    "vocoder_name": "nsf-hifigan",
-    "speech_encoder": "vec768l12",
-    "speaker_embedding": false,
-    "vol_embedding": true
-  },
-  "spk": {
-    "tomori": 0
-  }
-}

models/tomori/tomori.pth

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:5e915470eed2322969ae9dfadb51393f2f5744092690c11149de0ed65f04413f
-size 209238367

models/tomori/tomori_index.pkl

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:2a069b9dfacaec64cbfa500576134535cac59b9fc08087f52c2cbeae90db9076
-size 189669700