Update app.py

app.py

@@ -1,13 +1,13 @@
 import gradio as gr
 from huggingface_hub import InferenceClient
 import difflib
+from transformers import pipeline  # Import transformers to load the speech-to-text model
 
 # Load Hugging Face Inference client
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
-# Load the speech-to-text model from Hugging Face
-s2t = gr.Interface.load('huggingface/facebook/s2t-medium-librispeech-asr')
-
+# Load the speech-to-text model using transformers pipeline
+s2t_model = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-large-960h-lv60-self")
 
 def generate_text_with_huggingface(system_message, max_tokens, temperature, top_p):
     """
@@ -57,7 +57,7 @@ def transcribe_and_feedback(audio, system_message, max_tokens, temperature, top_
     reference_text = generate_text_with_huggingface(system_message, max_tokens, temperature, top_p)
 
     # Transcribe the audio using the speech-to-text model
-    transcription = s2t(audio)
+    transcription = s2t_model(audio)["text"]
 
     # Provide pronunciation feedback based on the transcription and the generated text
     feedback = pronunciation_feedback(transcription, reference_text)