| |
| """new.ipynb |
| |
| Automatically generated by Colab. |
| |
| Original file is located at |
| https://colab.research.google.com/drive/10l5M_bqlqlmx8isz58qrPCp9YwDOKjiC |
| """ |
| """" |
| from re import S |
| !pip install Streamlit |
| !pip install whisper |
| !pip install pydub |
| !pip install mapbox |
| !pip install openai-whisper pydub streamlit requests geopy |
| !pip install streamlit-webrtc |
| """" |
| |
| from google.colab import userdata |
| userdata.get('secret') |
| |
| import streamlit as st |
| import requests |
| from geopy.geocoders import Nominatim |
| import whisper |
| import tempfile |
| from pydub import AudioSegment |
| from io import BytesIO |
| import base64 |
| import folium |
| import requests |
| from mapbox import Geocoder |
| import folium |
| |
| |
| # Set your Hugging Face API URL and API key |
| API_URL = "https://api-inference.huggingface.co/models/dmis-lab/biobert-base-cased-v1.1" |
| headers = {"Authorization": f"secret"} |
| |
| # Initialize Whisper model |
| whisper_model = whisper.load_model("base") |
| |
| # Initialize Mapbox Geocoder with your API key |
| MAPBOX_API_KEY = "sk.eyJ1IjoibWFyaW9jYWxlYiIsImEiOiJjbHdkZWV3b2wwdnlxMmtwbnhqaHdnNzA3In0.iOhyIXlUFztYtc1iZhSMng" |
| geocoder = Geocoder(access_token=MAPBOX_API_KEY) |
| |
| # Function to query the Hugging Face model |
| def query(payload): |
| response = requests.post(API_URL, headers=headers, json=payload) |
| if response.status_code == 200: |
| return response.json() |
| else: |
| st.error(f"Error: Unable to fetch response from model (status code: {response.status_code})") |
| st.error(response.text) |
| return None |
| |
| # Function to find nearby clinics/pharmacies using Mapbox |
| def find_nearby_clinics(address): |
| response = geocoder.forward(address) |
| if response.status_code == 200: |
| features = response.json()['features'] |
| if features: |
| location = features[0]['center'] |
| return location |
| else: |
| st.error("No locations found") |
| return None |
| else: |
| st.error("Error: Unable to fetch location data") |
| return None |
| |
| # Function to transcribe audio to text using Whisper |
| def transcribe_audio(audio_bytes): |
| with tempfile.NamedTemporaryFile(delete=False, suffix=".wav") as temp_audio_file: |
| audio = AudioSegment.from_file(BytesIO(audio_bytes), format="wav") |
| audio.export(temp_audio_file.name, format="wav") |
| result = whisper_model.transcribe(temp_audio_file.name) |
| return result["text"] |
| |
| # Main function to create the Streamlit app |
| def main(): |
| st.title("Healthcare Companion") |
| st.write("Welcome to Healthcare Companion! Your personal healthcare guide.") |
| |
| st.header("Speak Your Symptoms") |
| st.write("Press the button and speak your symptoms clearly.") |
| |
| # JavaScript code to capture audio |
| js_code = """ |
| async function recordAudio() { |
| const stream = await navigator.mediaDevices.getUserMedia({ audio: true }); |
| const mediaRecorder = new MediaRecorder(stream); |
| let audioChunks = []; |
|
|
| mediaRecorder.ondataavailable = event => { |
| audioChunks.push(event.data); |
| }; |
|
|
| mediaRecorder.onstop = async () => { |
| const audioBlob = new Blob(audioChunks, { type: 'audio/wav' }); |
| const audioBuffer = await audioBlob.arrayBuffer(); |
| const audioBase64 = arrayBufferToBase64(audioBuffer); |
| document.getElementById('audio_data').value = audioBase64; |
| document.getElementById('audio_form').submit(); |
| }; |
|
|
| mediaRecorder.start(); |
| setTimeout(() => mediaRecorder.stop(), 5000); // Record for 5 seconds |
|
|
| function arrayBufferToBase64(buffer) { |
| let binary = ''; |
| const bytes = new Uint8Array(buffer); |
| const len = bytes.byteLength; |
| for (let i = 0; i < len; i++) { |
| binary += String.fromCharCode(bytes[i]); |
| } |
| return window.btoa(binary); |
| } |
| } |
|
|
| recordAudio(); |
| """ |
| |
| # Placeholder for audio data |
| #st_js_code = streamlit_js_eval(js_code, key="record_audio") |
| |
| # Form to receive audio data from JavaScript |
| with st.form("audio_form", clear_on_submit=True): |
| audio_data = st.text_input("audio_data", type="default") |
| submit_button = st.form_submit_button("Submit") |
| |
| if submit_button and audio_data: |
| audio_bytes = BytesIO(base64.b64decode(audio_data)) |
| symptoms = transcribe_audio(audio_bytes) |
| st.write(f"**Transcribed symptoms:** {symptoms}") |
| |
| if 'symptoms' in locals() and symptoms: |
| st.header("Symptom Checker") |
| st.write("Enter your symptoms below for advice.") |
| |
| context = """ |
| This is a healthcare question and answer platform. The following text contains typical symptoms, treatments, and medical conditions commonly asked about in healthcare settings. |
| For example, symptoms of COVID-19 include fever, dry cough, and tiredness. Treatment options for hypertension include lifestyle changes and medications. The platform is designed to assist with general medical inquiries. |
| """ |
| payload = {"inputs": {"question": symptoms, "context": context}} |
| result = query(payload) |
| if result: |
| st.write("**Medical Advice:**") |
| answer = result.get('answer', "Sorry, I don't have information on that.") |
| st.write(answer) |
| |
| st.header("Locate Nearest Pharmacy/Clinic") |
| st.write("Enter your address to find the nearest pharmacy or clinic.") |
| |
| address = st.text_input("Enter your address here:") |
| if address: |
| location = find_nearby_clinics(address) |
| if location: |
| map_ = folium.Map(location=[location[1], location[0]], zoom_start=15) |
| folium.Marker([location[1], location[0]], popup="Your Location").add_to(map_) |
| st_folium(map_, width=700, height=500) |
| st.write(f"**Nearby Clinics/Pharmacies (Coordinates):** {location}") |
| |
| st.write("Providing reliable healthcare guidance and information. Please consult a healthcare professional for medical emergencies.") |
| |
| if __name__ == "__main__": |
| main() |
| |
| |
