Sound event detection system and method

Abstract

The invention discloses a sound event detection system and method. The sound event detection system comprises a voice activity detection subsystem, a database and a sound event detection subsystem. The voice activity detection subsystem comprises a voice receiving module, a feature acquisition module and a first judgment module. The voice receiving module receives an original voice signal, the feature capturing module captures a plurality of features from the original voice signal, and the first judgment module executes a first classification process to judge whether the features conform to a starting voice. The database is used for storing the captured features. The sound event detection subsystem comprises a second judgment module and a function response module. The second determination module executes a second classification process to determine whether the features conform to at least one of a plurality of predetermined voices. The function response module executes at least one of a plurality of functions corresponding to the predetermined voice determined to conform to the predetermined voice.

Description

technical field [0001] The present invention relates to a sound event detection system and method, in particular to a sound event detection system and method that can save storage space and computing power consumption. Background technique [0002] Existing audio wake-up applications are mostly used to detect certain "events", such as voice commands or sound events (crying, glass breaking, etc.), and trigger response actions, such as sending command data to the cloud or sounding an alarm signal. [0003] Audio wake-up applications are mostly implemented with an "Always-on" system. In other words, the detection system is always "monitoring" the ambient sound and collecting the required voice signals. A system that is always on is very power hungry. In order to effectively control power consumption, most devices use voice activity detection (Voice activity detection, VAD) to filter out most of the invalid voice signals, so as to avoid too many entering the stage of acoustic e...

AI Technical Summary

Problems solved by technology

However, in the VAD and AED stages mentioned above, the power consumption of feature extraction becomes very important

Images