Automatic Speech Recognition , a service used to recognize the user's voice and translate the voice into text. Its goal is to convert the vocabulary content in human speech into computer-readable input, so that the device has the ability to 'hear' and realize human-computer interaction.