Speech recognition method, application thereof in smart classroom and storage medium

The invention discloses a speech recognition method, an application thereof in a smart classroom and a storage medium. The speech recognition method comprises the following steps: S01, determining a master control sound acquisition area and a background sound acquisition area; s02, acquiring sound information greater than a preset decibel in real time, and generating sound acquisition information; s03, sound information smaller than a preset decibel is acquired according to a preset time frequency, and background sound information is generated; s04, acquiring sound acquisition information, performing instruction positioning and cutting on the sound acquisition information according to a preset condition, and outputting voice instruction information; s05, acquiring voice instruction information, and converting the voice instruction information into text information; s06, matching the character information with a preset instruction library, and outputting a matching result; s07, determining to execute the instruction or output the optional instruction according to the matching result; s08, outputting the determined execution instruction; wherein the sound acquisition information is acquired by a main control sound acquisition area; the background sound information is acquired by the background sound acquisition area; the scheme is reliable in implementation, accurate in identification, high in efficiency and humanized in use.