Spaces:
Sleeping
Sleeping
shamimjony1000
commited on
Commit
β’
c8b7640
1
Parent(s):
e52cb87
Update voice_handler.py
Browse files- voice_handler.py +25 -60
voice_handler.py
CHANGED
@@ -1,8 +1,6 @@
|
|
1 |
import speech_recognition as sr
|
2 |
-
import streamlit as st
|
3 |
-
from typing import Optional, Tuple
|
4 |
import platform
|
5 |
-
import
|
6 |
|
7 |
class VoiceHandler:
|
8 |
def __init__(self):
|
@@ -11,38 +9,25 @@ class VoiceHandler:
|
|
11 |
self.recognizer.dynamic_energy_threshold = True
|
12 |
self.recognizer.pause_threshold = 0.8
|
13 |
self.permission_granted = False
|
14 |
-
|
15 |
-
# Initialize microphone without device selection
|
16 |
-
self.mic = sr.Microphone()
|
17 |
|
18 |
-
def check_microphone_access(self) ->
|
19 |
-
"""Check if microphone is accessible
|
20 |
try:
|
21 |
-
with
|
22 |
-
# Quick test to verify microphone access
|
23 |
self.recognizer.adjust_for_ambient_noise(source, duration=0.1)
|
24 |
-
return True
|
25 |
-
except (OSError, AttributeError)
|
26 |
-
return False
|
27 |
-
Could not access microphone. Please:
|
28 |
-
1. Click the lock icon in your browser's address bar
|
29 |
-
2. Allow microphone access for this site
|
30 |
-
3. Refresh the page after granting access
|
31 |
-
4. Ensure your microphone is properly connected
|
32 |
-
"""
|
33 |
-
except sr.RequestError as e:
|
34 |
-
return False, f"Speech recognition service error: {str(e)}"
|
35 |
-
except Exception as e:
|
36 |
-
return False, f"Unexpected error: {str(e)}"
|
37 |
|
38 |
-
def request_permissions(self) ->
|
39 |
"""Request microphone permissions from the browser"""
|
40 |
-
|
41 |
-
|
42 |
-
|
43 |
-
|
44 |
-
|
45 |
-
|
|
|
46 |
|
47 |
def listen_for_voice(self, language: str = "mixed") -> str:
|
48 |
"""
|
@@ -52,47 +37,27 @@ class VoiceHandler:
|
|
52 |
- "en-US" for English
|
53 |
- "mixed" for both Arabic and English
|
54 |
"""
|
55 |
-
if not self.permission_granted:
|
56 |
-
|
57 |
-
if not success:
|
58 |
-
st.error(message)
|
59 |
-
st.markdown("""
|
60 |
-
### π€ Troubleshooting Steps:
|
61 |
-
1. Check browser compatibility (Chrome/Firefox/Edge recommended)
|
62 |
-
2. Verify microphone connection
|
63 |
-
3. Check system sound settings
|
64 |
-
4. Try a different microphone if available
|
65 |
-
5. Restart browser if needed
|
66 |
-
""")
|
67 |
-
return message
|
68 |
|
69 |
try:
|
70 |
-
with
|
71 |
-
|
72 |
self.recognizer.adjust_for_ambient_noise(source, duration=1)
|
73 |
-
|
74 |
-
st.info("π€ Listening... Speak now!")
|
75 |
audio = self.recognizer.listen(source, timeout=5, phrase_time_limit=10)
|
|
|
76 |
|
77 |
-
st.info("Processing speech...")
|
78 |
return self._process_audio(audio, language)
|
79 |
|
80 |
except sr.RequestError as e:
|
81 |
-
|
82 |
-
st.error(error_msg)
|
83 |
-
return error_msg
|
84 |
except sr.UnknownValueError:
|
85 |
-
|
86 |
-
st.warning(error_msg)
|
87 |
-
return error_msg
|
88 |
except sr.WaitTimeoutError:
|
89 |
-
|
90 |
-
st.warning(error_msg)
|
91 |
-
return error_msg
|
92 |
except Exception as e:
|
93 |
-
|
94 |
-
st.error(error_msg)
|
95 |
-
return error_msg
|
96 |
|
97 |
def _process_audio(self, audio, language: str) -> str:
|
98 |
"""Process audio input and convert to text"""
|
|
|
1 |
import speech_recognition as sr
|
|
|
|
|
2 |
import platform
|
3 |
+
from typing import Optional
|
4 |
|
5 |
class VoiceHandler:
|
6 |
def __init__(self):
|
|
|
9 |
self.recognizer.dynamic_energy_threshold = True
|
10 |
self.recognizer.pause_threshold = 0.8
|
11 |
self.permission_granted = False
|
|
|
|
|
|
|
12 |
|
13 |
+
def check_microphone_access(self) -> bool:
|
14 |
+
"""Check if microphone is accessible"""
|
15 |
try:
|
16 |
+
with sr.Microphone() as source:
|
|
|
17 |
self.recognizer.adjust_for_ambient_noise(source, duration=0.1)
|
18 |
+
return True
|
19 |
+
except (OSError, AttributeError, sr.RequestError):
|
20 |
+
return False
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
|
22 |
+
def request_permissions(self) -> bool:
|
23 |
"""Request microphone permissions from the browser"""
|
24 |
+
try:
|
25 |
+
if self.check_microphone_access():
|
26 |
+
self.permission_granted = True
|
27 |
+
return True
|
28 |
+
return False
|
29 |
+
except Exception:
|
30 |
+
return False
|
31 |
|
32 |
def listen_for_voice(self, language: str = "mixed") -> str:
|
33 |
"""
|
|
|
37 |
- "en-US" for English
|
38 |
- "mixed" for both Arabic and English
|
39 |
"""
|
40 |
+
if not self.permission_granted and not self.request_permissions():
|
41 |
+
return "Error: Please grant microphone permissions to use voice input."
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
42 |
|
43 |
try:
|
44 |
+
with sr.Microphone() as source:
|
45 |
+
print("Adjusting for ambient noise...")
|
46 |
self.recognizer.adjust_for_ambient_noise(source, duration=1)
|
47 |
+
print("Listening...")
|
|
|
48 |
audio = self.recognizer.listen(source, timeout=5, phrase_time_limit=10)
|
49 |
+
print("Processing speech...")
|
50 |
|
|
|
51 |
return self._process_audio(audio, language)
|
52 |
|
53 |
except sr.RequestError as e:
|
54 |
+
return f"Could not request results from speech service: {str(e)}"
|
|
|
|
|
55 |
except sr.UnknownValueError:
|
56 |
+
return "Could not understand audio. Please speak clearly and try again."
|
|
|
|
|
57 |
except sr.WaitTimeoutError:
|
58 |
+
return "Listening timed out. Please try again."
|
|
|
|
|
59 |
except Exception as e:
|
60 |
+
return f"Error: {str(e)}"
|
|
|
|
|
61 |
|
62 |
def _process_audio(self, audio, language: str) -> str:
|
63 |
"""Process audio input and convert to text"""
|