Update app.py

app.py

@@ -6,7 +6,6 @@ from huggingface_hub import snapshot_download
 import numpy as np
 from scipy.io import wavfile
 
-
 model_ids = [
     'suno/bark',
 ]
@@ -49,12 +48,22 @@ def infer(prompt, input_wav_file):
 
     # cloning a speaker.
     # It assumes that you have a speaker file in `bark_voices/speaker_n/speaker.wav` or `bark_voices/speaker_n/speaker.npz`
-    output_dict = model.synthesize(text, config, speaker_id=f"{file_name}", voice_dirs="bark_voices/")
+    output_dict = model.synthesize(
+        text, 
+        config, 
+        speaker_id=f"{file_name}", 
+        voice_dirs="bark_voices/"
+    )
+    
     print(output_dict)
 
     sample_rate = 24000  # Replace with the actual sample rate
 
-    wavfile.write('output.wav', sample_rate, output_dict['wav'])
+    wavfile.write(
+        'output.wav', 
+        sample_rate, 
+        output_dict['wav']
+    )
 
     # List all the files and subdirectories in the given directory
     contents = os.listdir(f"bark_voices/{file_name}")
@@ -63,13 +72,27 @@ def infer(prompt, input_wav_file):
     for item in contents:
         print(item)   
     
-    return "output.wav"
-
-gr.Interface(fn=infer, 
-             inputs=[gr.Textbox(label="Text to speech prompt"), 
-                     gr.Audio(
-                         label="WAV voice to clone", 
-                         type="filepath",
-                         source="upload")],
-             outputs=[gr.Audio()], 
-             title="Instant Voice Cloning").launch()
+    return "output.wav", f"bark_voices/{file_name}/{content[1]}"
+
+gr.Interface(
+    fn=infer, 
+    inputs=[
+        gr.Textbox(
+            label="Text to speech prompt"
+        ), 
+        gr.Audio(
+            label="WAV voice to clone", 
+            type="filepath",
+            source="upload"
+        )
+    ],
+    outputs=[
+        gr.Audio(
+            label="Text to speech output"
+        ),
+        gr.File(
+            label=".npz file"
+        )
+    ], 
+    title="Instant Voice Cloning"
+).launch()