Upload folder using huggingface_hub

app.py

@@ -14,12 +14,11 @@ asr_pipe = pipeline("automatic-speech-recognition", model="openai/whisper-large-
 model_id = "Sandiago21/speecht5_finetuned_mozilla_foundation_common_voice_13_german"  # update with your model id
 # pipe = pipeline("automatic-speech-recognition", model=model_id)
 model = SpeechT5ForTextToSpeech.from_pretrained(model_id)
+processor = SpeechT5Processor.from_pretrained(model_id)
 vocoder = SpeechT5HifiGan.from_pretrained("microsoft/speecht5_hifigan")
 embeddings_dataset = load_dataset("Matthijs/cmu-arctic-xvectors", split="validation")
 speaker_embeddings = torch.tensor(embeddings_dataset[7440]["xvector"]).unsqueeze(0)
 
-processor = SpeechT5Processor.from_pretrained(model_id)
-
 replacements = [    
     ("Ä", "E"),
     ("Æ", "E"),
@@ -112,8 +111,8 @@ 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 Greek. Demo uses OpenAI's [Whisper Large v2](https://huggingface.co/openai/whisper-large-v2) model for speech translation, and [Sandiago21/speecht5_finetuned_mozilla_foundation_common_voice_13_german](https://huggingface.co/Sandiago21/speecht5_finetuned_mozilla_foundation_common_voice_13_german) checkpoint for text-to-speech, which is based on Microsoft's
-[SpeechT5 TTS](https://huggingface.co/microsoft/speecht5_tts) model for text-to-speech, fine-tuned in Greek Audio dataset:
+Demo for cascaded speech-to-speech translation (STST), mapping from source speech in any language to target speech in German. Demo uses OpenAI's [Whisper Large v2](https://huggingface.co/openai/whisper-large-v2) model for speech translation, and [Sandiago21/speecht5_finetuned_mozilla_foundation_common_voice_13_german](https://huggingface.co/Sandiago21/speecht5_finetuned_mozilla_foundation_common_voice_13_german) checkpoint for text-to-speech, which is based on Microsoft's
+[SpeechT5 TTS](https://huggingface.co/microsoft/speecht5_tts) model for text-to-speech, fine-tuned in German Audio dataset:
 ![Cascaded STST](https://huggingface.co/datasets/huggingface-course/audio-course-images/resolve/main/s2st_cascaded.png "Diagram of cascaded speech to speech translation")
 """