feat: add functionality to upload and refresh custom voice samples in demo interface

app.py

@@ -2,7 +2,8 @@
 VibeVoice Gradio Demo - High-Quality Dialogue Generation Interface with Streaming Support
 """
 
-import argparse
+import argparse, os
+import tempfile
 import torch
 import gradio as gr
 
@@ -44,6 +45,14 @@ def create_demo_interface(demo_instance: VibeVoiceDemo):
         with gr.Row():
             # Left column - Settings
             with gr.Column(scale=1, elem_classes="settings-card"):
+
+                def process_and_refresh_voices(uploaded_files: list[tempfile._TemporaryFileWrapper]):
+                    if not uploaded_files: return [gr.update() for _ in speaker_selections] + [None]
+                    for f in uploaded_files: 
+                        demo_instance.available_voices[os.path.basename(f.name)] = f.name
+                    new_choices = list(demo_instance.available_voices.keys())
+                    return [gr.update(choices=new_choices) for _ in speaker_selections] + [None]
+
                 gr.Markdown("### 🎛️ **Podcast Settings**")
                 
                 # Number of speakers
@@ -75,7 +84,13 @@ def create_demo_interface(demo_instance: VibeVoiceDemo):
                         elem_classes="speaker-item"
                     )
                     speaker_selections.append(speaker)
-                
+
+                with gr.Accordion("🎤 Upload Custom Voices", open=False):
+                    upload_audio = gr.File(label="Upload Voice Samples", file_count="multiple", file_types=["audio"])
+                    process_upload_btn = gr.Button("Add Uploaded Voices to Speaker Selection")
+                process_upload_btn.click(fn=process_and_refresh_voices, inputs=upload_audio, outputs=speaker_selections + [upload_audio])
+
+
                 # Advanced settings
                 gr.Markdown("### ⚙️ **Advanced Settings**")