An Open-set Recognition and Few-Shot Learning Dataset for Audio Event Classification in Domestic Environments

sounds
Download Naranjo-Alcazar2021

General information

Label Value Description
Name An Open-set Recognition and Few-Shot Learning Dataset for Audio Event Classification in Domestic Environments Full dataset name
ID sounds/fsl_osr Datalist id for external indexing
Abbreviation FSL-OSR Official dataset abbreviation, e.g. one used in the original paper
Provider Visualfy, Universitat de Valencia
Year 2021 Dataset release year
Modalities Audio Data modalities included in the dataset
Collection name FSL-OSR Common name for all related datasets, used to group datasets coming from same source
Research domain Open-Set Few-shot learning Tagging Related domains, e.g., Scenes, Mobile devices, Audio-visual, Open set, Ambient noise, Unlabelled, Multiple sensors, SED, SELD, Tagging, FL, Strong annotation, Weak annotation, Unlabelled, Multi-annotator
License Creative Commons, CC BY 4.0
Download Download (142.1 MB)
Citation [Naranjo-Alcazar2021] An Open-Set Recognition and Few-Shot Learning Dataset for Audio Event Classification in Domestic Environments

Audio

Label Value Description

Data

Data type Audio Possible values: Audio | Features
File format
File format type Constant Possible values: Constant | Variable
File format wav Possible value: wav | aiff | flac | mp3 | aac | ogg
Lossy compression No is audio compressed in a lossy manner
Bit rate 16 Bit depth of audio, possible values: 8 | 16 | 24 | 32
Sampling rate (kHz) 16 kHz Sampling rate in kHz, possible values: 8 | 16 | 22.05 | 32 | 44.1 | 48
Channels
Setup Mono Possible values: Mono | Stereo | Binaural | Ambisonic | Array | Multi-Channel | Variable
Number of channels 1
Material
Source Original Possible values: Original | Youtube | Freesound | Online | Crowdsourced | [Dataset name]

Content

Content type Isolated Possible values: Freefield | Synthetic | Isolated
Scene Constant Is the scene class constant for single recording, possible values: Constant | Variable

Recording

Files

Count 1360 files Total number of files
Total duration (minutes) 90.666 min Total duration of the dataset in minutes
File length Constant Characterization of the file lengths, possible values: Constant | Quasi-constant | Variable
File length (seconds) 4 sec Approximate length of files

Meta

Label Value Description
Types Tag List of meta data types provided for the data, possible values: Event, Tag, Scene, Caption, Geolocation, Spatial location, Annotator, Timestamp, Presence, Proximity, etc.

Event

Classes 11 Number of event classes
Classes True Possible values: True | False | Almost
Annotation
Type Weak Possible values: Strong | Weak | Location | None
Source Experts Possible values: Experts | Crowdsourced | Synthetic | Metadata | Automatic
Annotations per item 1 How many annotations there are available per item (possible multi-annotator setup)
Labelled amount (%) 100 % Percentage of all data, amount of data which is labelled
Validated amount (%) 100 % Percentage of all data, amount of data which is validated by human
Strong annotations amount (%) 0 % Percentage of all data, amount of data which has strong annotations
Overlapping event instances No
Labeling
Hierarchical No
Instance
Count 1360 Count of all event instances in the dataset
Average instances per class 123.63 Average per class instance count

Cross-validation setup

Label Value Description
Provided Yes

Baseline

Label Value Description
Download Download Link to baseline system source code
Citation [Naranjo-Alcazar2021] Paper to cite for the baseline