import streamlit as st import requests # Function to send the audio file to the Hugging Face Whisper API def query(file_data, my_key): API_URL = "https://api-inference.huggingface.co/models/openai/whisper-large-v3-turbo" headers = {"Authorization": f"Bearer {my_key}"} try: response = requests.post(API_URL, headers=headers, files={'file': file_data}) return response.json() except requests.exceptions.RequestException as e: return {"error": str(e)} # Streamlit UI elements st.title("Whisper Transcription App") st.write("Upload a .wav, .mp3, or .flac audio file, and get the transcription.") # Get the user's Hugging Face API key my_key = st.text_input('Enter your Hugging Face API Key', type='password') # File uploader for audio files uploaded_files = st.file_uploader("Choose an audio file", type=["mp3", "wav", "flac"], accept_multiple_files=True) if my_key: # Proceed only if the API key is provided if uploaded_files: results = {} for uploaded_file in uploaded_files: st.write(f"Processing file: {uploaded_file.name}") # Check the MIME type of the uploaded file file_type = uploaded_file.type st.write(f"File type: {file_type}") # Validate file type (must be one of the supported types) if file_type not in ["audio/mpeg", "audio/wav", "audio/flac"]: st.write(f"Unsupported file type: {file_type}. Please upload an MP3, WAV, or FLAC file.") continue # Send the file to the Hugging Face API for transcription output = query(uploaded_file, my_key) # Store and display the result results[uploaded_file.name] = output # Show the transcription results for all uploaded files st.write("Results:") for file, result in results.items(): st.write(f"**Results for {file}:**") st.json(result) else: st.write("Please upload an audio file to transcribe.") else: st.write("Please enter your Hugging Face API key to proceed.")