Database Schema

Database schema diagram

Tables

Tables

Table Name

Description

annotation

The primary table containing all the annotations

job

Metadata pertinent to the various annotation jobs performed by human analysts or machines

model

Models and algorithms employed to process acoustic data

file

Audio files

deployment

Hydrophone deployments

taxonomy

Taxonomies used for annotating the data

label

Sound-source and sound-type combinations allowed for labelling sounds

file_job_relation

Mapping of many-to-many relationship between audio files and annotation jobs

Annotation

Annotation table

Name

Type

Description

Example/Allowed Values

id

INTEGER

Index

job_id

INTEGER

Annotation job index

deployment_id

INTEGER

Deployment index

file_id

INTEGER

Audio file index

label_id

INTEGER

Label index of confident source/type assignment

tentative_label_id

INTEGER

Label index of tentative source/type assignment

num_files

INTEGER

Number of audio files spanned by the annotation

typically 1

file_id_list

JSON

List of audio file indices (only relevant when num_files > 1)

[3, 17]

start_utc

TEXT

UTC start time in ISO8601 format (YYYY-MM-DD HH:MM:SS.SSS)

duration_ms

INTEGER

Duration of the acoustic signal in milliseconds

start_ms

INTEGER

Start time of acoustic signal in milliseconds relative to the beginning of the audio file

freq_min_hz

INTEGER

Lower frequency bound of acoustic signal in Hz

110

freq_max_hz

INTEGER

Upper frequency bound of acoustic signal in Hz

320

channel

INTEGER

Stereo channel number

0,1,…

granularity

TEXT

Granularity of the annotation

call,window,batch,file,encounter

machine_prediction

JSON

Model classification scores

{“class”:[“KW”, “HW”], “score”:[0.8, 0.2]}

comments

TEXT

Additional observations

Etc.