Systems and methods to detect end of speech in a noisy environment

A system configured to detect end of speech in a noisy environment, the system comprising:

• one or more processors configured by machine-readable instructions to:
• start recording of sound by an audio section of a client computing platform in the noisy environment, the sound recorded by the audio section including spoken input from a user and background noise from the noisy environment;
• perform speech recognition on the sound recorded by the audio section to determine the spoken input;
• detect, after starting the recording of the sound, a beginning of the spoken input from the user;
• determine a local maximum of a volume level of the sound which occurs during a period from the beginning of the spoken input from the user to a completion of a given number of words or sentences spoken from the beginning;
• dynamically determine a volume level threshold for a remainder of the recording of the sound by the audio section based on the local maximum of the volume level of the sound within the given number of the words or the sentences spoken from the beginning of the spoken input from the user;
• determine, in an ongoing manner as the sound is continued to be recorded by the audio section, the volume level of the sound;
• detect, responsive to a determination that the volume level of the sound is less than the volume level threshold, an end of the spoken input from the user;
• and
• terminate, after a predetermined amount of time following detection of the end of the spoken input, the recording of the sound by the audio section.