Update app.py

app.py

@@ -49,7 +49,7 @@ def predict(text, speaker):
     else:
         speaker_embedding = np.load(speaker_embeddings[speaker[:3]])
 
-    speaker_embedding = torch.tensor(speaker_embedding).unsqueeze(0)
+    speaker_embedding = torch.tensor([1, 512]).unsqueeze(0)
 
     speech = model.generate_speech(input_ids, speaker_embedding, vocoder=vocoder)
 
@@ -111,15 +111,6 @@ gr.Interface(
     fn=predict,
     inputs=[
         gr.Text(label="Input Text"),
-        gr.Radio(label="Speaker", choices=[
-            "BDL (male)",
-            "CLB (female)",
-            "KSP (male)",
-            "RMS (male)",
-            "SLT (female)",
-            "Surprise Me!"
-        ],
-        value="BDL (male)"),
     ],
     outputs=[
         gr.Audio(label="Generated Speech", type="numpy"),