dcase_ Util tutorial

An audio scene recognition based on python Assembly of dcase_util, Is English , So it is translated into Chinese here , For reference .

dcase_util The document describes the detection and classification of acoustic scenes and events （DCASE） A collection of utilities created . These utilities were originally designed for DCASE Challenge the baseline system （2016 and 2017） Created , And bundled into a separate library to allow it to be reused in other research projects .

The main goal of these utilities is to simplify the research code , Make it easier to read , Easier to maintain . Most implemented utilities are related to audio datasets ： Dealing with metadata and other structured data in various forms , And provide standardized use for audio data sets from various sources API.

dcase_util The concepts contained in are ：

Dataset： A set of audio signals and their associated reference labels .
Container： Class stores data and is based on the type of data to store , Provide meaningful and clear data access .
Metadata： Data annotation .
Repository： Container to store multiple data containers .
Encoder： Used to convert data from a different type to another type .
Processor： Provide a uniform for processing data API Class .
ProcessingChain： Connecting multiple data processors in a chain , Allows complex processing of building data .
It's being built python After the environment , Use the following pip Command to install dcase_util.

---

---

---

Reference material ：
dcase_util course