app.py

from flask import Flask, render_template, request, jsonify
import torch
from transformers import pipeline
from gtts import gTTS
import os
import re

app = Flask(__name__)

# Load Whisper Model for English Transcription
device = "cuda" if torch.cuda.is_available() else "cpu"
asr_model = pipeline("automatic-speech-recognition", model="openai/whisper-base", device=0 if device == "cuda" else -1)

# Function to generate audio prompts
def generate_audio_prompt(text, filename):
    tts = gTTS(text=text, lang="en")
    tts.save(os.path.join("static", filename))

# Generate audio prompts
prompts = {
    "welcome": "Welcome to Biryani Hub.",
    "ask_name": "Tell me your name.",
    "ask_email": "Please provide your email address.",
    "thank_you": "Thank you for registration."
}

for key, text in prompts.items():
    generate_audio_prompt(text, f"{key}.mp3")

# Clean transcribed text to allow only English letters, numbers, and basic punctuation
def clean_transcription(text):
    return re.sub(r"[^a-zA-Z0-9@.\s]", "", text)

@app.route("/")
def index():
    return render_template("index.html")

@app.route("/transcribe", methods=["POST"])
def transcribe():
    if "audio" not in request.files:
        return jsonify({"error": "No audio file provided"}), 400

    audio_file = request.files["audio"]
    audio_path = os.path.join("static", "temp.wav")
    audio_file.save(audio_path)

    try:
        # Transcribe audio to text
        result = asr_model(audio_path, generate_kwargs={"language": "en"})
        transcribed_text = clean_transcription(result["text"])
        return jsonify({"text": transcribed_text})
    except Exception as e:
        return jsonify({"error": str(e)}), 500

if __name__ == "__main__":
    app.run(host="0.0.0.0", port=5000, debug=True)