Update app.py

app.py

@@ -6,7 +6,7 @@ from datasets import load_dataset
 from transformers import SpeechT5ForTextToSpeech, SpeechT5HifiGan, SpeechT5Processor, pipeline
 #from transformers import pipeline
 #from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
-#from transformers import BarkModel, BarkProcessor
+from transformers import BarkModel, BarkProcessor
 
 device = "cuda:0" if torch.cuda.is_available() else "cpu"
 
@@ -14,8 +14,11 @@ device = "cuda:0" if torch.cuda.is_available() else "cpu"
 asr_pipe = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-base-100h", device=device)
 
 # load text-to-speech checkpoint and speaker embeddings
-processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
-model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts").to(device)
+#processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
+#model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts").to(device)
+
+model = BarkModel.from_pretrained("suno/bark-small")
+processor = BarkProcessor.from_pretrained("suno/bark-small")
 vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan").to(device)
 
 #model = BarkModel.from_pretrained("suno/bark-small")