app.py

from dotenv import load_dotenv
import os
import gradio as gr
from brain_of_doctor import load_image, analyze_image_with_query
from voice_of_the_patient import record_audio, transcription_with_groq
from voice_of_the_doctor import (
    text_to_speech_with_gtts_new_autoplay,
)
from deep_translator import GoogleTranslator
from gtts import gTTS

load_dotenv()

system_prompt = """You have to act as a professional doctor, i know you are not but this is for learning purpose. 
What's in this image?. Do you find anything wrong with it medically? 
If you make a differential, suggest some remedies for them. Donot add any numbers or special characters in 
your response. Your response should be in one long paragraph. Also always answer as if you are answering to a real person.
Donot say 'In the image I see' but say 'With what I see, I think you have ....'
Dont respond as an AI model in markdown, your answer should mimic that of an actual doctor not an AI bot, 
Keep your answer concise (max 2 sentences). No preamble, start your answer right away please"""

# ✅ Helper for Urdu TTS (using gTTS)
def tts_with_language_support(text, output_file="final.wav", lang="en"):
    try:
        tts = gTTS(text=text, lang=lang)
        tts.save(output_file)
        return output_file
    except Exception as e:
        print("TTS generation failed:", e)
        return None


def process_inputs(audio_filePath, image_filePath, language):
    # Speech → text
    speech_to_text_output = transcription_with_groq(
        GROQ_API_KEY= os.getenv("GROQ_API_KEY"),
        audio_filePath=audio_filePath,
        sst_model="whisper-large-v3"
    )

    # Image analysis
    if image_filePath:
        doctor_response_en = analyze_image_with_query(
            query=system_prompt + speech_to_text_output,
            encoded_image=load_image(image_filePath),
            model="meta-llama/llama-4-scout-17b-16e-instruct"
        )
    else:
        doctor_response_en = "No image provided for me to analyze"

    # Translation + voice selection
    if language == "Urdu":
        # Translate to Urdu
        doctor_response = GoogleTranslator(source='en', target='ur').translate(doctor_response_en)
        # Generate Urdu voice with gTTS
        output_audio = tts_with_language_support(text=doctor_response, output_file="final.wav", lang="ur")
    else:
        # English voice via ElevenLabs
        doctor_response = doctor_response_en
        output_audio = text_to_speech_with_gtts_new_autoplay(
            input_text=doctor_response,
            output_file="final.wav"
        )
        output_audio= "final.wav"

    return speech_to_text_output, doctor_response, output_audio


# ✅ Gradio Interface
iface = gr.Interface(
    fn=process_inputs,
    inputs=[
        gr.Audio(sources=["microphone"], type="filepath", label="Patient's Voice"),
        gr.Image(type="filepath", label="Upload Medical Image"),
        gr.Radio(choices=["English", "Urdu"], value="English", label="Select Language")
    ],
    outputs=[
        gr.Textbox(label="Speech to Text"),
        gr.Textbox(label="Doctor's Response"),
        gr.Audio(label="Doctor's Voice (Auto Play)", autoplay=True, type="filepath")
    ],
    title="AI Doctor — Developed by Danish Khan"
)

iface.launch(debug=True)