Update app.py

app.py

@@ -27,11 +27,8 @@ def create_vc_fn(tgt_sr, net_g, vc, if_f0, file_index, file_big_npy):
         index_rate
     ):
         try:
-            if args.files:
-                audio, sr = librosa.load(input_audio, sr=16000, mono=True)
-            else:
-                if input_audio is None:
-                    return "You need to upload an audio", None
+            # Check if input_audio is a tuple (Gradio input)
+            if isinstance(input_audio, tuple):
                 sampling_rate, audio = input_audio
                 duration = audio.shape[0] / sampling_rate
                 if duration > 10000000:
@@ -41,6 +38,9 @@ def create_vc_fn(tgt_sr, net_g, vc, if_f0, file_index, file_big_npy):
                     audio = librosa.to_mono(audio.transpose(1, 0))
                 if sampling_rate != 16000:
                     audio = librosa.resample(audio, orig_sr=sampling_rate, target_sr=16000)
+            else:  # Assume it's a file path
+                audio, sr = librosa.load(input_audio, sr=16000, mono=True)
+
             times = [0, 0, 0]
             f0_up_key = int(f0_up_key)
             audio_opt = vc.pipeline(
@@ -66,6 +66,7 @@ def create_vc_fn(tgt_sr, net_g, vc, if_f0, file_index, file_big_npy):
             return info, (None, None)
     return vc_fn
 
+
 def load_hubert():
     global hubert_model
     models, _, _ = checkpoint_utils.load_model_ensemble_and_task(
@@ -154,4 +155,16 @@ if __name__ == '__main__':
                             vc_output1 = gr.Textbox(label="Output Message")
                             vc_output2 = gr.Audio(label="Output Audio")
                 vc_submit.click(vc_fn, [vc_input, vc_transpose, vc_f0method, vc_index_ratio], [vc_output1, vc_output2])
-        app.queue(concurrency_count=1, max_size=20, api_open=args.api).launch(share=args.share)
+        app.queue(concurrency_count=1, max_size=20, api_open=args.api).launch(share=args.share)
+iface = gr.Interface(
+    fn=create_vc_fn(tgt_sr, net_g, vc, if_f0, file_index, file_big_npy), 
+    inputs=[
+        gr.inputs.Audio(source="microphone", type="numpy", sample_rate=16000),
+        gr.inputs.Slider(minimum=-12, maximum=12, step=1, default=0),
+        gr.inputs.Radio(["world", "dio"], label="F0 method"),
+        gr.inputs.Slider(minimum=0, maximum=1, step=0.01, default=0.5)
+    ], 
+    outputs="text",
+)
+
+iface.launch()