Update app.py

app.py

@@ -23,8 +23,8 @@ processor = SpeechT5Processor.from_pretrained("sanchit-gandhi/speecht5_tts_vox_n
 model = SpeechT5ForTextToSpeech.from_pretrained("sanchit-gandhi/speecht5_tts_vox_nl").to(device)
 vocoder = SpeechT5HifiGan.from_pretrained("sanchit-gandhi/speecht5_tts_vox_nl").to(device)
 
-# embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
-# speaker_embeddings = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)
+embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
+speaker_embeddings = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)
 
 spk_model_name = "speechbrain/spkrec-xvect-voxceleb"
 
@@ -49,7 +49,7 @@ speaker_embeddings_list = []
 
 for i, data in enumerate(iter(dataset_nl)):
     # print(i)
-    if(i > 5):
+    if(i > 16):
       break
     data_list.append(data)
     # data = next(iter(dataset_nl))
@@ -60,7 +60,7 @@ for i, data in enumerate(iter(dataset_nl)):
     speaker_embeddings_list.append(speaker_embeddings)
 
 
-speaker_embeddings = speaker_embeddings_list[4]
+speaker_embeddings = speaker_embeddings_list[11]
     
 def translate(audio):
     # outputs = asr_pipe(audio, max_new_tokens=256, generate_kwargs={"task": "translate"})