Update app.py

app.py

@@ -43,6 +43,18 @@ def synthesize_speech(text):
     return (16000, speech)
 
 def create_speaker_embedding(waveform):
+
+    import os
+    from speechbrain.inference.speaker import EncoderClassifier
+    
+    spk_model_name = "speechbrain/spkrec-xvect-voxceleb"
+    
+    device = "cuda" if torch.cuda.is_available() else "cpu"
+    speaker_model = EncoderClassifier.from_hparams(
+        source=spk_model_name, 
+        run_opts={"device": device}, 
+        savedir=os.path.join("/tmp", spk_model_name),
+    )
     with torch.no_grad():
         speaker_embeddings = speaker_model.encode_batch(torch.tensor(waveform))
         speaker_embeddings = torch.nn.functional.normalize(speaker_embeddings, dim=2)