init

app.py

@@ -40,7 +40,7 @@ def get_text(text, hps):
     return text_norm
 
 
-hps_ms = utils.get_hparams_from_file("configs/vctk_base.json")
+hps_ms = utils.get_hparams_from_file("configs/nyarumul.json")
 net_g_ms = SynthesizerTrn(
     len(symbols),
     hps_ms.data.filter_length // 2 + 1,
@@ -53,7 +53,7 @@ featureInput = FeatureInput(hps_ms.data.sampling_rate, hps_ms.data.hop_length)
 
 hubert = torch.hub.load("bshall/hubert:main", "hubert_soft")
 
-_ = utils.load_checkpoint("G_312000.pth", net_g_ms, None)
+_ = utils.load_checkpoint("nyarumodel.pth", net_g_ms, None)
 
 def vc_fn(input_audio,vc_transform):
     if input_audio is None:
@@ -95,7 +95,7 @@ with app:
     with gr.Tabs():
         with gr.TabItem("Basic"):
             vc_input3 = gr.Audio(label="Input Audio (30s limitation)")
-            vc_transform = gr.Number(label="transform",value=1.0)
+            vc_transform = gr.Number(label="变调数量",value=1)
             vc_submit = gr.Button("Convert", variant="primary")
             vc_output1 = gr.Textbox(label="Output Message")
             vc_output2 = gr.Audio(label="Output Audio")

nyarumodel.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50330133cf1807c7c96daf2e424c4e84895500c15117f52b9b81208d5c3eae26
+size 221105477