A request to recognize speech from captured audio content, such as audio from the device’s microphone.