|
from flask import Flask, render_template_string, request, jsonify |
|
import speech_recognition as sr |
|
from tempfile import NamedTemporaryFile |
|
import os |
|
import ffmpeg |
|
import logging |
|
from werkzeug.exceptions import BadRequest |
|
|
|
|
|
app = Flask(__name__) |
|
|
|
|
|
logging.basicConfig(level=logging.INFO) |
|
|
|
|
|
html_code = """ |
|
<!DOCTYPE html> |
|
<html lang="en"> |
|
<head> |
|
<meta charset="UTF-8"> |
|
<meta name="viewport" content="width=device-width, initial-scale=1.0"> |
|
<title>AI Dining Assistant</title> |
|
<style> |
|
body { |
|
font-family: Arial, sans-serif; |
|
background-color: #f4f4f9; |
|
display: flex; |
|
flex-direction: column; |
|
align-items: center; |
|
justify-content: center; |
|
height: 100vh; |
|
margin: 0; |
|
} |
|
h1 { |
|
color: #333; |
|
} |
|
.mic-button { |
|
width: 80px; |
|
height: 80px; |
|
border-radius: 50%; |
|
background-color: #007bff; |
|
color: white; |
|
font-size: 24px; |
|
border: none; |
|
display: flex; |
|
align-items: center; |
|
justify-content: center; |
|
cursor: pointer; |
|
box-shadow: 0px 4px 6px rgba(0, 0, 0, 0.1); |
|
transition: background-color 0.3s; |
|
} |
|
.mic-button:hover { |
|
background-color: #0056b3; |
|
} |
|
.status { |
|
margin-top: 20px; |
|
font-size: 18px; |
|
color: #666; |
|
} |
|
.listening { |
|
color: green; |
|
font-weight: bold; |
|
} |
|
.response { |
|
margin-top: 20px; |
|
padding: 10px; |
|
background-color: #fff; |
|
border: 1px solid #ddd; |
|
border-radius: 5px; |
|
box-shadow: 0px 4px 6px rgba(0, 0, 0, 0.1); |
|
width: 300px; |
|
text-align: center; |
|
} |
|
</style> |
|
</head> |
|
<body> |
|
<h1>AI Dining Assistant</h1> |
|
<button class="mic-button" id="mic-button">🎤</button> |
|
<div class="status" id="status">Press the mic button to start the conversation...</div> |
|
<div class="response" id="response" style="display: none;">Response will appear here...</div> |
|
<script> |
|
const micButton = document.getElementById('mic-button'); |
|
const status = document.getElementById('status'); |
|
const response = document.getElementById('response'); |
|
let mediaRecorder; |
|
let audioChunks = []; |
|
let isConversationActive = false; |
|
|
|
micButton.addEventListener('click', () => { |
|
if (!isConversationActive) { |
|
isConversationActive = true; |
|
startConversation(); |
|
} |
|
}); |
|
|
|
function startConversation() { |
|
status.textContent = 'Listening...'; |
|
startListening(); |
|
} |
|
|
|
function startListening() { |
|
navigator.mediaDevices.getUserMedia({ audio: true }).then(stream => { |
|
mediaRecorder = new MediaRecorder(stream, { mimeType: 'audio/webm;codecs=opus' }); |
|
mediaRecorder.start(); |
|
audioChunks = []; |
|
mediaRecorder.ondataavailable = event => audioChunks.push(event.data); |
|
mediaRecorder.onstop = async () => { |
|
const audioBlob = new Blob(audioChunks, { type: 'audio/webm' }); |
|
const formData = new FormData(); |
|
formData.append('audio', audioBlob); |
|
|
|
status.textContent = 'Processing...'; |
|
try { |
|
const result = await fetch('/process-audio', { method: 'POST', body: formData }); |
|
const data = await result.json(); |
|
response.textContent = data.response; |
|
response.style.display = 'block'; |
|
|
|
try { |
|
const utterance = new SpeechSynthesisUtterance(data.response); |
|
speechSynthesis.speak(utterance); |
|
|
|
utterance.onend = () => { |
|
console.log("Speech synthesis completed."); |
|
}; |
|
|
|
utterance.onerror = (e) => { |
|
console.error("Speech synthesis error:", e.error); |
|
status.textContent = 'Error with speech output.'; |
|
}; |
|
} catch (speechError) { |
|
console.error("Speech synthesis not supported or failed:", speechError); |
|
response.textContent = "Speech output unavailable. Please check your browser."; |
|
} |
|
|
|
if (data.response.includes("Goodbye")) { |
|
status.textContent = 'Conversation ended. Press the mic button to start again.'; |
|
isConversationActive = false; |
|
} else { |
|
status.textContent = 'Listening...'; |
|
setTimeout(startListening, 1000); // Continue listening |
|
} |
|
} catch (error) { |
|
response.textContent = 'Error occurred. Please try again.'; |
|
response.style.display = 'block'; |
|
status.textContent = 'Press the mic button to restart the conversation.'; |
|
isConversationActive = false; |
|
} |
|
}; |
|
setTimeout(() => mediaRecorder.stop(), 5000); // Stop recording after 5 seconds |
|
}).catch(() => { |
|
status.textContent = 'Microphone access denied.'; |
|
isConversationActive = false; |
|
}); |
|
} |
|
</script> |
|
</body> |
|
</html> |
|
""" |
|
|
|
@app.route('/') |
|
def index(): |
|
return render_template_string(html_code) |
|
|
|
@app.route('/process-audio', methods=['POST']) |
|
def process_audio(): |
|
try: |
|
|
|
audio_file = request.files.get('audio') |
|
if not audio_file: |
|
raise BadRequest("No audio file provided.") |
|
|
|
temp_file = NamedTemporaryFile(delete=False, suffix=".webm") |
|
audio_file.save(temp_file.name) |
|
logging.info(f"Saved input audio to {temp_file.name}") |
|
|
|
if os.path.getsize(temp_file.name) == 0: |
|
raise BadRequest("Uploaded audio file is empty.") |
|
|
|
|
|
converted_file = NamedTemporaryFile(delete=False, suffix=".wav") |
|
try: |
|
ffmpeg.input(temp_file.name).output( |
|
converted_file.name, acodec='pcm_s16le', ac=1, ar='16000' |
|
).run(overwrite_output=True) |
|
except Exception as ffmpeg_error: |
|
logging.error(f"FFmpeg conversion error: {str(ffmpeg_error)}") |
|
return jsonify({"response": "Audio conversion failed. Please try again."}) |
|
|
|
logging.info(f"Converted audio saved to {converted_file.name}") |
|
|
|
|
|
recognizer = sr.Recognizer() |
|
with sr.AudioFile(converted_file.name) as source: |
|
audio_data = recognizer.record(source) |
|
command = recognizer.recognize_google(audio_data) |
|
logging.info(f"Recognized command: {command}") |
|
response = process_command(command) |
|
|
|
return jsonify({"response": response}) |
|
|
|
except BadRequest as br: |
|
logging.error(f"Bad request error: {br}") |
|
return jsonify({"response": f"Bad Request: {str(br)}"}) |
|
|
|
except Exception as e: |
|
logging.error(f"Error processing audio: {e}") |
|
return jsonify({"response": f"An error occurred: {str(e)}"}) |
|
|
|
finally: |
|
|
|
try: |
|
if os.path.exists(temp_file.name): |
|
os.unlink(temp_file.name) |
|
if os.path.exists(converted_file.name): |
|
os.unlink(converted_file.name) |
|
except Exception as cleanup_error: |
|
logging.error(f"Error cleaning up files: {cleanup_error}") |
|
|
|
def process_command(command): |
|
"""Process the user's voice command and return a response.""" |
|
command = command.lower() |
|
if "menu" in command: |
|
return ( |
|
"Here is our menu: " |
|
"South Indian dishes include Idli, Dosa, Vada, Pongal, Biryani, and Sambar Rice. " |
|
"North Indian dishes include Butter Chicken, Paneer Butter Masala, Naan, Dal Makhani, Chole Bhature, and Rajma Chawal. " |
|
"What would you like to order?" |
|
) |
|
elif "order" in command: |
|
return "Your order has been placed. Would you like anything else?" |
|
elif "no" in command or "nothing" in command: |
|
return "Goodbye! Thank you for using AI Dining Assistant." |
|
return "Sorry, I didn't understand your request. Please ask about the menu or place an order." |
|
|
|
if __name__ == "__main__": |
|
app.run(host="0.0.0.0", port=7860) |
|
|
