change app for whisper testing

app.py

@@ -1,32 +1,34 @@
-# import streamlit as st
-#
-# x = st.slider('Select a value')
-# st.write(x, 'squared is', x * x)
-
-
 import streamlit as st
-from transformers import pipeline
+import whisper
+import tempfile
+from pydub import AudioSegment
+
+# Load the Whisper model
+model = whisper.load_model("base")
 
-st.write("Starting the app")
+st.title("Voice Recognition App using Whisper")
 
-# Load a pre-trained pipeline for text generation
-generator = pipeline('text-generation', model='gpt2')
+st.write("Upload an audio file and the Whisper model will transcribe it to text.")
 
-st.title('Simple Hugging Face Space with Streamlit')
+# File uploader for audio file
+uploaded_file = st.file_uploader("Choose an audio file", type=["wav", "mp3", "m4a"])
 
-# Text input widget
-input_text = st.text_area('Enter your text here:', '')
+if uploaded_file is not None:
+    # Save the uploaded file temporarily
+    with tempfile.NamedTemporaryFile(delete=False) as temp_file:
+        temp_file.write(uploaded_file.read())
+        temp_file_path = temp_file.name
 
-if st.button('Generate'):
-    if input_text:
-        # Generate text using the model
-        outputs = generator(input_text, max_length=50, num_return_sequences=1)
-        st.write('Generated Text:')
-        st.write(outputs[0]['generated_text'])
-    else:
-        st.write('Please enter some text to generate.')
+    # Convert audio file to a format supported by Whisper (if necessary)
+    audio = AudioSegment.from_file(temp_file_path)
+    temp_wav_path = tempfile.mktemp(suffix=".wav")
+    audio.export(temp_wav_path, format="wav")
 
-# if __name__ == '__main__':
-#     st.title('ACR')
+    st.audio(uploaded_file, format="audio/wav")
 
+    st.write("Transcribing audio...")
 
+    # Transcribe audio using Whisper model
+    result = model.transcribe(temp_wav_path)
+    st.write("Transcription:")
+    st.write(result["text"])