Spaces:
Sleeping
Sleeping
| import speech_recognition as sr | |
| import streamlit as st | |
| from typing import Optional, Tuple | |
| class VoiceHandler: | |
| def __init__(self): | |
| self.recognizer = sr.Recognizer() | |
| self.recognizer.energy_threshold = 4000 | |
| self.recognizer.dynamic_energy_threshold = True | |
| self.recognizer.pause_threshold = 0.8 | |
| self.permission_granted = False | |
| def check_microphone_access(self) -> Tuple[bool, str]: | |
| """Check if microphone is accessible and return status with message""" | |
| try: | |
| with sr.Microphone() as source: | |
| self.recognizer.adjust_for_ambient_noise(source, duration=0.1) | |
| return True, "Microphone access granted" | |
| except (OSError, AttributeError) as e: | |
| return False, "Could not access microphone. Please check your browser settings and microphone connection." | |
| except sr.RequestError as e: | |
| return False, f"Error accessing speech service: {str(e)}" | |
| except Exception as e: | |
| return False, f"Unexpected error: {str(e)}" | |
| def request_permissions(self) -> Tuple[bool, str]: | |
| """Request microphone permissions from the browser""" | |
| success, message = self.check_microphone_access() | |
| if success: | |
| self.permission_granted = True | |
| return success, message | |
| def listen_for_voice(self, language: str = "mixed") -> str: | |
| """ | |
| Listen for voice input in specified language. | |
| language can be: | |
| - "ar-SA" for Arabic | |
| - "en-US" for English | |
| - "mixed" for both Arabic and English | |
| """ | |
| # Check permissions first | |
| if not self.permission_granted: | |
| success, message = self.request_permissions() | |
| if not success: | |
| st.error(message) | |
| st.markdown(""" | |
| ### Troubleshooting Steps: | |
| 1. Click the lock/info icon in your browser's address bar | |
| 2. Ensure microphone access is allowed for this site | |
| 3. Check if your microphone is properly connected | |
| 4. Try refreshing the page | |
| 5. If using Chrome, go to Settings > Privacy and Security > Site Settings > Microphone | |
| """) | |
| return message | |
| try: | |
| with sr.Microphone() as source: | |
| st.info("Adjusting for ambient noise... Please wait.") | |
| self.recognizer.adjust_for_ambient_noise(source, duration=1) | |
| st.info("🎤 Listening... Speak now!") | |
| audio = self.recognizer.listen(source, timeout=5, phrase_time_limit=10) | |
| st.info("Processing speech...") | |
| return self._process_audio(audio, language) | |
| except sr.RequestError as e: | |
| error_msg = f"Could not request results from speech service: {str(e)}" | |
| st.error(error_msg) | |
| return error_msg | |
| except sr.UnknownValueError: | |
| error_msg = "Could not understand audio. Please speak clearly and try again." | |
| st.warning(error_msg) | |
| return error_msg | |
| except sr.WaitTimeoutError: | |
| error_msg = "Listening timed out. Please try again." | |
| st.warning(error_msg) | |
| return error_msg | |
| except Exception as e: | |
| error_msg = f"Error: {str(e)}" | |
| st.error(error_msg) | |
| return error_msg | |
| def _process_audio(self, audio, language: str) -> str: | |
| """Process audio input and convert to text""" | |
| if language in ["ar-SA", "mixed"]: | |
| try: | |
| return self.recognizer.recognize_google(audio, language="ar-SA") | |
| except sr.UnknownValueError: | |
| if language == "mixed": | |
| return self.recognizer.recognize_google(audio, language="en-US") | |
| raise | |
| return self.recognizer.recognize_google(audio, language="en-US") |