update

inference/gradio/gradio_settings.yaml

@@ -5,7 +5,7 @@ article: |
   Link to <a href='https://github.com/Rongjiehuang/GenerSpeech' style='color:blue;' target='_blank\'>Github REPO</a>
 example_inputs:
   - |-
-    the invention of movable metal letters in the middle of the fifteenth century may justly be considered as the invention of the art of printing.
+    the invention of movable metal letters in the middle of the fifteenth century may justly be considered as the invention of the art of printing. | assets/0011_001570.wav
   - |-
     produced the block books, which were the immediate predecessors of the true printed book,
 inference_cls: inference.GenerSpeech.GenerSpeechInfer

inference/gradio/infer.py

@@ -51,10 +51,16 @@ class GradioInfer:
         infer_cls = self.inference_cls
         self.infer_ins: BaseTTSInfer = infer_cls(hp)
         example_inputs = self.example_inputs
+
+        for i in range(len(example_inputs)):
+            text, ref_audio = example_inputs[i].split('|')
+            print('text: ', text, 'ref_audio:', ref_audio)
+            example_inputs[i] = [text, reference_audio]
+
         iface = gr.Interface(fn=self.greet,
                              inputs=[
-                             Textbox(lines=10, placeholder=None, default=example_inputs[0], label="input text"),
-                             Audio(label="reference audio"),
+                             Textbox(lines=10, placeholder=None, default=example_inputs[0][0], label="input text"),
+                             Textbox(lines=10, placeholder=None, default=example_inputs[0][1], label="reference audio"),
                              ],
                              outputs="audio",
                              allow_flagging="never",