Update app.py

app.py

@@ -20,13 +20,13 @@ speaker_embeddings = {
 }
 
 
-def predict(text, speaker):
+def predict(text):
     if len(text.strip()) == 0:
         return (16000, np.zeros(0).astype(np.int16))
 
     inputs = processor(text=text, return_tensors="pt")
 
-    speaker_embedding = np.random.default_rng().uniform(-0.20,0.20,512)
+    speaker_embedding = np.float32(np.random.default_rng().uniform(-0.20,0.20,512))
     speaker_embedding = torch.tensor(speaker_embedding).unsqueeze(0)
 
     speech = model.generate_speech(inputs["input_ids"], speaker_embedding, vocoder=vocoder)