Spaces:
Build error
Build error
DSatishchandra
commited on
Commit
•
d67403c
1
Parent(s):
e7f0157
Update order_assistant.py
Browse files- order_assistant.py +70 -52
order_assistant.py
CHANGED
@@ -1,68 +1,86 @@
|
|
1 |
-
import
|
2 |
import random
|
|
|
3 |
import speech_recognition as sr
|
4 |
import edge_tts
|
5 |
-
import asyncio
|
6 |
import tempfile
|
|
|
7 |
|
8 |
# Load the food menu from menu.json
|
9 |
-
|
10 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
11 |
|
12 |
-
|
13 |
-
def
|
14 |
-
|
15 |
-
return
|
16 |
-
return "No suggestion available"
|
17 |
|
18 |
-
# Speech-to-Text function
|
19 |
-
def recognize_speech_from_mic():
|
20 |
-
recognizer = sr.Recognizer()
|
21 |
-
with sr.Microphone() as source:
|
22 |
-
print("Say something!")
|
23 |
-
audio = recognizer.listen(source)
|
24 |
-
try:
|
25 |
-
return recognizer.recognize_google(audio) # Use Google Speech Recognition
|
26 |
-
except sr.UnknownValueError:
|
27 |
-
return "Sorry I didn't catch that"
|
28 |
-
except sr.RequestError:
|
29 |
-
return "Sorry, I'm having trouble reaching the service"
|
30 |
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
tmp_path = tmp_file.name
|
36 |
-
await communicate.save(tmp_path)
|
37 |
-
return tmp_path
|
38 |
|
39 |
-
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
audio_path = await text_to_speech(greeting_text)
|
44 |
|
45 |
-
|
46 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
47 |
|
48 |
-
|
49 |
-
|
|
|
|
|
|
|
|
|
50 |
|
51 |
-
|
52 |
-
|
|
|
53 |
|
54 |
-
|
55 |
-
|
|
|
56 |
|
57 |
-
|
58 |
-
|
|
|
59 |
|
60 |
-
|
61 |
-
|
62 |
-
|
63 |
-
|
64 |
-
|
65 |
-
|
66 |
-
|
67 |
-
|
68 |
-
|
|
|
1 |
+
import base64
|
2 |
import random
|
3 |
+
import numpy as np
|
4 |
import speech_recognition as sr
|
5 |
import edge_tts
|
|
|
6 |
import tempfile
|
7 |
+
import json
|
8 |
|
9 |
# Load the food menu from menu.json
|
10 |
+
def load_food_menu():
|
11 |
+
with open("menu.json") as f:
|
12 |
+
return json.load(f)
|
13 |
+
|
14 |
+
food_menu = load_food_menu()
|
15 |
+
|
16 |
+
class AudioProcessor:
|
17 |
+
@staticmethod
|
18 |
+
def encode_audio(data, sample_rate):
|
19 |
+
encoded = base64.b64encode(data.tobytes()).decode("UTF-8")
|
20 |
+
return {
|
21 |
+
"realtimeInput": {
|
22 |
+
"mediaChunks": [
|
23 |
+
{
|
24 |
+
"mimeType": f"audio/pcm;rate={sample_rate}",
|
25 |
+
"data": encoded,
|
26 |
+
}
|
27 |
+
],
|
28 |
+
},
|
29 |
+
}
|
30 |
|
31 |
+
@staticmethod
|
32 |
+
def process_audio_response(data):
|
33 |
+
audio_data = base64.b64decode(data)
|
34 |
+
return np.frombuffer(audio_data, dtype=np.int16)
|
|
|
35 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
36 |
|
37 |
+
class OrderAssistant:
|
38 |
+
def __init__(self, food_menu):
|
39 |
+
self.food_menu = food_menu
|
40 |
+
self.audio_processor = AudioProcessor()
|
|
|
|
|
|
|
41 |
|
42 |
+
def get_food_suggestion(self, food_type, dietary_preference):
|
43 |
+
if food_type in self.food_menu:
|
44 |
+
return random.choice(self.food_menu[food_type].get(dietary_preference, []))
|
45 |
+
return "Sorry, we don't have that option."
|
|
|
46 |
|
47 |
+
def recognize_speech_from_mic(self):
|
48 |
+
recognizer = sr.Recognizer()
|
49 |
+
with sr.Microphone() as source:
|
50 |
+
print("Say something!")
|
51 |
+
audio = recognizer.listen(source)
|
52 |
+
try:
|
53 |
+
return recognizer.recognize_google(audio) # Google Speech Recognition
|
54 |
+
except sr.UnknownValueError:
|
55 |
+
return "Sorry I didn't catch that"
|
56 |
+
except sr.RequestError:
|
57 |
+
return "Sorry, I'm having trouble reaching the service"
|
58 |
|
59 |
+
async def text_to_speech(self, text):
|
60 |
+
communicate = edge_tts.Communicate(text, "en-US-AriaNeural", rate="0%", pitch="0Hz")
|
61 |
+
with tempfile.NamedTemporaryFile(delete=False, suffix=".mp3") as tmp_file:
|
62 |
+
tmp_path = tmp_file.name
|
63 |
+
await communicate.save(tmp_path)
|
64 |
+
return tmp_path
|
65 |
|
66 |
+
async def food_order_assistant(self, audio_input, food_type):
|
67 |
+
greeting_text = "Welcome to the restaurant! What would you like to order today? Are you looking for Vegan, Halal, or Guilt-Free options?"
|
68 |
+
audio_path = await self.text_to_speech(greeting_text)
|
69 |
|
70 |
+
dietary_preference = self.recognize_speech_from_mic().lower()
|
71 |
+
if dietary_preference not in ["vegan", "halal", "guilt-free"]:
|
72 |
+
dietary_preference = "vegan" # Default to Vegan if unrecognized
|
73 |
|
74 |
+
suggestion = self.get_food_suggestion(food_type, dietary_preference)
|
75 |
+
suggestion_text = f"I suggest you try {suggestion}. Does that sound good?"
|
76 |
+
suggestion_audio = await self.text_to_speech(suggestion_text)
|
77 |
|
78 |
+
confirmation = self.recognize_speech_from_mic().lower()
|
79 |
+
if "yes" in confirmation:
|
80 |
+
confirmation_text = f"Your order for {suggestion} has been confirmed. Sending the order to the kitchen."
|
81 |
+
confirmation_audio = await self.text_to_speech(confirmation_text)
|
82 |
+
return confirmation_audio, f"Confirmed order: {suggestion}"
|
83 |
+
else:
|
84 |
+
cancellation_text = "Okay, let's try again."
|
85 |
+
cancellation_audio = await self.text_to_speech(cancellation_text)
|
86 |
+
return cancellation_audio, "Order not confirmed"
|