VRS (Voice Recognition Software)
A powerful voice recognition library made in python. This combines several libraries in order to achieve a "Voice Assistant" ready library.
Features
-
Works both offline and offline (Online will utilize the google recognizer from SpeechRecognitionI)
-
Realtime speech recognition through the use of VOSK
-
Use any wakeword/hotword you want. Some wakewords will utilize pvporcupine for better accuracy while some will rely on VOSK's realtime capability.
-
Allows for preprocessing before performing final recognition (i.e., remove bg noise, increase certain frequencies, etc.)