Upload app.py

app.py

@@ -1,4 +1,3 @@
-
 import re
 import gradio as gr
 import torch
@@ -92,7 +91,7 @@ def selection(speaker):
         spk = 8
         return spk
 
-def infer(text,speaker_id):
+def infer(text,speaker_id, n_scale= 0.667,n_scale_w = 0.8, l_scale = 1 ):
     text = clean_text(text)
     speaker_id = int(selection(speaker_id))
     dev = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
@@ -102,7 +101,7 @@ def infer(text,speaker_id):
         x_tst = stn_tst.unsqueeze(0).to(dev)
         x_tst_lengths = torch.LongTensor([stn_tst.size(0)]).to(dev)
         sid = torch.LongTensor([speaker_id]).to(dev)
-        audio = net_g_ms.infer(x_tst, x_tst_lengths, sid=sid, noise_scale=0.667, noise_scale_w=0.8, length_scale=1)[0][
+        audio = net_g_ms.infer(x_tst, x_tst_lengths, sid=sid, noise_scale=n_scale, noise_scale_w=n_scale_w, length_scale=l_scale)[0][
             0, 0].data.cpu().float().numpy()
     return (hps_ms.data.sampling_rate, audio)
 
@@ -110,10 +109,16 @@ idols = ["南小鸟","园田海未","小泉花阳","星空凛","东条希","矢
 app = gr.Blocks()
 with app:
     with gr.Tabs():
-        with gr.TabItem("面板"):
-            tts_input1 = gr.TextArea(label="请输入纯中文或纯日文", value="大家好，今天给大家来点想看的东西啊")
-            speaker1 = gr.Dropdown(label="选择说话人",choices=idols, value="高坂穗乃果", interactive=True)
+
+        with gr.TabItem("Basic"):
+
+            tts_input1 = gr.TextArea(label="请输入纯中文或纯日文", value="大家好")
+            para_input1 = gr.Slider(minimum= 0.01,maximum=1.0,label="更改噪声比例", value=0.667)
+            para_input2 = gr.Slider(minimum= 0.01,maximum=1.0,label="更改噪声偏差", value=0.8)
+            para_input3 = gr.Slider(minimum= 0.1,maximum=10,label="更改时间比例", value=1)
             tts_submit = gr.Button("Generate", variant="primary")
+            speaker1 = gr.Dropdown(label="选择说话人",choices=idols, value="高坂穗乃果", interactive=True)
             tts_output2 = gr.Audio(label="Output")
-            tts_submit.click(infer, [tts_input1,speaker1], [tts_output2])
+
+            tts_submit.click(infer, [tts_input1,speaker1,para_input1,para_input2,para_input3], [tts_output2])
     app.launch()