Update app.py

app.py

@@ -15,14 +15,14 @@ asr_pipe = pipeline("automatic-speech-recognition", model="openai/whisper-base",
 #processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
 #Use own TTS Model
 #processor = SpeechT5Processor.from_pretrained("jasonl1/speecht5_finetuned_voxpopuli_fi")
-processor = SpeechT5Processor.from_pretrained("sanchit-gandhi/speecht5_tts_vox_nl", ignore_mismatched_sizes=True)
+processor = SpeechT5Processor.from_pretrained("sanchit-gandhi/speecht5_tts_vox_nl")
 
 #model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts").to(device)
 #Use own TTS Model
 #model = SpeechT5ForTextToSpeech.from_pretrained("jasonl1/speecht5_finetuned_voxpopuli_fi")
-model = SpeechT5ForTextToSpeech.from_pretrained("sanchit-gandhi/speecht5_tts_vox_nl",ignore_mismatched_sizes=True)
+model = SpeechT5ForTextToSpeech.from_pretrained("sanchit-gandhi/speecht5_tts_vox_nl")
 
-vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan",ignore_mismatched_sizes=True).to(device)
+vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan").to(device)
 
 embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
 speaker_embeddings = torch.tensor(embeddings_dataset[7306]["xvector"]).unsqueeze(0)