Update app.py

app.py

@@ -4,6 +4,8 @@ import torch
 from datasets import load_dataset
 
 from transformers import SpeechT5ForTextToSpeech, SpeechT5HifiGan, SpeechT5Processor, pipeline
+from transformers import BarkModel, BarkProcessor
+
 
 
 device = "cuda:0" if torch.cuda.is_available() else "cpu"
@@ -12,10 +14,13 @@ device = "cuda:0" if torch.cuda.is_available() else "cpu"
 asr_pipe = pipeline("automatic-speech-recognition", model="openai/whisper-base", device=device)
 
 # load text-to-speech checkpoint and speaker embeddings
-processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
+# processor = SpeechT5Processor.from_pretrained("microsoft/speecht5_tts")
+
+barkmodel = BarkModel.from_pretrained("suno/bark")
+barkprocessor = BarkProcessor.from_pretrained("suno/bark")
 
-model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts").to(device)
-vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan").to(device)
+# model = SpeechT5ForTextToSpeech.from_pretrained("microsoft/speecht5_tts").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)
@@ -27,9 +32,9 @@ def translate(audio):
 
 
 def synthesise(text):
-    inputs = processor(text=text, return_tensors="pt")
-    speech = model.generate_speech(inputs["input_ids"].to(device), speaker_embeddings.to(device), vocoder=vocoder)
-    return speech.cpu()
+    inputs = barkprocessor(text=[text], voice_preset="v2/it_speaker_4",return_tensors="pt")
+    speech = barkmodel.generate(**inputs, do_sample=True)
+    return speech
 
 
 def speech_to_speech_translation(audio):
@@ -41,8 +46,8 @@ def speech_to_speech_translation(audio):
 
 title = "Cascaded STST"
 description = """
-Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in Italian. Demo uses OpenAI's [Whisper Base](https://huggingface.co/openai/whisper-base) model for speech translation, and Microsoft's
-[SpeechT5 TTS](https://huggingface.co/microsoft/speecht5_tts) model for text-to-speech:
+Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in Italian. Demo uses OpenAI's [Whisper Base](https://huggingface.co/openai/whisper-base) model for speech translation, and Suno's
+[Bark](https://huggingface.co/suno/bark) model for text-to-speech:
 
 ![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")
 """