Launching a campaign of annotations on Zooniverse with ChildProject

Lucas Gautheron 2d318904ad Mise à jour de 'datacite.yml' 8 months ago
.datalad 3c425c0519 [DATALAD] new dataset 10 months ago
annotations 60c7f88082 going big 9 months ago
classifications 52cba13f0f more data 10 months ago
extra bb1a19cb7e retrieving and matching classifications 10 months ago
samples 15b16a50c8 initial data 10 months ago
vandam-data @ 1dcc4650d7 4dca2ead01 [DATALAD] Recorded changes 10 months ago
.gitattributes 2f9bbb2965 updated git attributes file 10 months ago
.gitmodules dd1df5abe3 [DATALAD] modified subdataset properties 10 months ago
LICENSE 72b94e89c4 License 8 months ago af9d3a3a91 tiny change 9 months ago
datacite.yml 2d318904ad Mise à jour de 'datacite.yml' 8 months ago

Zooniverse campaign


The present repository showcases the organization of a Zooniverse campaign using ChildProject and DataLad. Zooniverse is a crowd-sourcing platform that may be used for large-scale annotation tasks. The ExElang book provides examples of research goals for which Zooniverse may be useful.

The present campaign requires citizens to listen to 500 ms audio clips and then to perform the following tasks:

  1. Decide whether they hear speech from either a Baby, a Child, an Adolescent, an Adult, or no speech.
  2. Guess the gender of Adolescent or Adult speakers.
  3. Classify the type of sound among four categories (Canonical, Non-Canonical, Laughing, Crying).


  1. We used DataLad to manage this campaign (installation instructions here).
  2. The primary dataset (containing the audio and the metadata) was included in this repository as a subdataset. It was structured according to ChildProject's format.
  3. ChildProject was used to generate the samples, to upload the audio chunks to zooniverse, and to retrieve the classification (installation instructions here).

This repository contains all the scripts that we used to implement this workflow. You are welcome to re-use this code and adapt it to your needs.

Repository structure

  • annotations contains annotations built from the classifications retrieved from Zooniverse.
  • classifications contains the classifications retrieved from Zooniverse.
  • samples contains the samples that were selected as well as the chunks generated from them.
  • vandam-data is a subdataset containing VanDam Daylong corpus, structed according to ChildProject's format — this is very important as it allows to use all the features of ChildProject that will be used next.

Preparing samples

The first step is to prepare the samples to be annotated; this includes choosing the portions of audio to annotate, and then extracting audio files from these portions. See this issue for examples of sampling strategies that have been used or considered. Many of them involve splitting the audio in short chunks in order to preserve privacy. However, this technique may not be suited for certain annotation tasks (e.g. for semantics).


Sampling consists in selecting which portions of the recordings should be annotated by humans. It can be done through using the samplers provided in the ChildProject package.

Here, we sample 50 vocalizations per recording among all those detected and attributed to the key child (CHI) or a female adult (FEM) by the Voice Type Classifier:

# download VTC annotations
datalad get vandam-data/annotations/vtc/converted

# sample random CHI and FEM vocalizations from these annotations
child-project sampler vandam-data samples/chi_fem/ random-vocalizations \
  --annotation-set vtc \
  --target-speaker-type CHI FEM \
  --sample-size 50 \
  --by recording_filename

The sampler produces a CSV dataframe as samples/chi_fem/segments_YYYYMMDD_HHMMSS.csv, e.g. samples/chi_fem/segments_20210716_184443.csv.

Preparing the chunks for Zooniverse

After the samples have been generated, they have to be extracted from the audio and uploaded to Zooniverse, which can be done with ChildProject's extract-chunks function. However, these samples may contain private information about the participants. Therefore, they cannot be shared as is on a public crowd-sourcing platform. We therefore configure extract-chunks to split these samples into 500ms chunks, which will be classified in random order, thus preventing the recovery of sensitive information by the contributors:

datalad get vandam-data/recordings/converted/standard

child-project zooniverse extract-chunks vandam-data \
 --keyword chi_fem \
 --chunks-length 500 \
 --segments samples/chi_fem/segments_20210716_184443.csv \
 --destination samples/chi_fem/chunks \
 --profile standard

This will extract the audio chunks into samples/chi_fem/chunks/chunks/ and write a metadata file into samples/chi_fem/chunks (in our case, as samples/chi_fem/chunks/chunks_20210716_191944.csv).

See ChildProject's documentation for more information about the Zooniverse pipeline.

Uploading audio chunks to Zooniverse

Once the chunks have been extracted, the next step is to upload them to Zooniverse. Note that due to quotas, it is recommended to upload only a few at time (e.g. 1000 per day).

You will need to provide the numerical id of your Zooniverse project. Instructions to create a Zooniverse project are available here.

You will also need to set Zooniverse credentials as environment variables:

export PROJECT_ID=14957
child-project zooniverse upload-chunks \
 --chunks samples/chi_fem/chunks/chunks_20210716_191944.csv \
 --project-id $PROJECT_ID \
 --set-name vandam_chi_fem \
 --amount 1000

This will display a message for each chunk:

uploading chunk BN32_010007.mp3 (25153080,25153580)
uploading chunk BN32_010007.mp3 (45016146,45016646)
uploading chunk BN32_010007.mp3 (46794141,46794641)
uploading chunk BN32_010007.mp3 (14107752,14108252)
uploading chunk BN32_010007.mp3 (35709983,35710483)
uploading chunk BN32_010007.mp3 (45433933,45434433)
uploading chunk BN32_010007.mp3 (35711483,35711983)
uploading chunk BN32_010007.mp3 (38737938,38738438)
uploading chunk BN32_010007.mp3 (24586156,24586656)
uploading chunk BN32_010007.mp3 (15556956,15557456)
uploading chunk BN32_010007.mp3 (28439601,28440101)
uploading chunk BN32_010007.mp3 (27317629,27318129)
uploading chunk BN32_010007.mp3 (38391252,38391752)

The subject set and its subjects (i.e. the chunks) now appears in the project:

Zooniverse subjects Zooniverse subjects

Retrieving classifications

The classifications performed by citizens on Zooniverse for this project can be retrieved with ChildProject's retrieve-classifications command:

child-project zooniverse retrieve-classifications \
  --destination classifications/classifications.csv \
  --project-id $PROJECT_ID
$ head classifications/classifications.csv 

The output contains no information related to the metadata of the input dataset, which is the desired behavior (Zooniverse should not store any data that might compromise the privacy of the participants) It also contains all the classifications for this project, including those for data outside of the current campaign. As a result, these data alone cannot be exploited and they have to be matched to the chunks metadata.

Matching classifications back to the metadata

Classifications can be matched to the original metadata (the recording from which the clips were extracted, the timestamps of the clips, etc.) manually, but it is possible to retrieve the classifications from Zooniverse and match them with their metadata at the same time with ChildProject:

child-project zooniverse retrieve-classifications \
  --destination classifications/classifications.csv \
  --project-id $PROJECT_ID \
  --chunks samples/chi_fem/chunks/chunks*.csv

Now, only relevant chunks are returned, and they are associated to all corresponding metadata:

id user_id subject_id task_id answer_id workflow_id answer index recording_filename onset offset segment_onset segment_offset wav mp3 date_extracted uploaded project_id subject_set zooniverse_id keyword
346474371 2202359.0 64210633 T1 4 17576 Junk 38 BN32_010007.mp3 24587156 24587656 24584902 24588410 BN32_010007_24587156_24587656.wav BN32_010007_24587156_24587656.mp3 2021-07-16 19:19:44 True 14957 vandam_chi_fem 64210633 chi_fem
346474611 2202359.0 64210607 T1 4 17576 Junk 30 BN32_010007.mp3 14108752 14109252 14107993 14109011 BN32_010007_14108752_14109252.wav BN32_010007_14108752_14109252.mp3 2021-07-16 19:19:44 True 14957 vandam_chi_fem 64210607 chi_fem
346474648 2202359.0 64210647 T1 4 17576 Junk 43 BN32_010007.mp3 4713603 4714103 4713263 4714944 BN32_010007_4713603_4714103.wav BN32_010007_4713603_4714103.mp3 2021-07-16 19:19:44 True 14957 vandam_chi_fem 64210647 chi_fem
346474683 2202359.0 64210890 T1 4 17576 Junk 125 BN32_010007.mp3 4713103 4713603 4713263 4714944 BN32_010007_4713103_4713603.wav BN32_010007_4713103_4713603.mp3 2021-07-16 19:19:44 True 14957 vandam_chi_fem 64210890 chi_fem
346474902 2202359.0 64210872 T1 4 17576 Junk 119 BN32_010007.mp3 39329854 39330354 39328511 39331697 BN32_010007_39329854_39330354.wav BN32_010007_39329854_39330354.mp3 2021-07-16 19:19:44 True 14957 vandam_chi_fem 64210872 chi_fem
346475101 2202359.0 64210613 T1 4 17576 Junk 32 BN32_010007.mp3 4149238 4149738 4148967 4149510 BN32_010007_4149238_4149738.wav BN32_010007_4149238_4149738.mp3 2021-07-16 19:19:44 True 14957 vandam_chi_fem 64210613 chi_fem
346475237 2202359.0 64210774 T1 4 17576 Junk 86 BN32_010007.mp3 17157819 17158319 17157511 17158127 BN32_010007_17157819_17158319.wav BN32_010007_17157819_17158319.mp3 2021-07-16 19:19:44 True 14957 vandam_chi_fem 64210774 chi_fem
346474475 2202359.0 64210728 T1 0 17576 Baby 70 BN32_010007.mp3 37978055 37978555 37978100 37979011 BN32_010007_37978055_37978555.wav BN32_010007_37978055_37978555.mp3 2021-07-16 19:19:44 True 14957 vandam_chi_fem 64210728 chi_fem
346474475 2202359.0 64210728 T3 1 17576 Non-Canonical 70 BN32_010007.mp3 37978055 37978555 37978100 37979011 BN32_010007_37978055_37978555.wav BN32_010007_37978055_37978555.mp3 2021-07-16 19:19:44 True 14957 vandam_chi_fem 64210728 chi_fem

Importing classifications into the source dataset

Once the classifications have been recovered, they can be used to enrich the source dataset with more annotations. While this step may depend a lot on the type of annotation that you are doing, this repository provides an example.

The feed-annotations script does just that. It can be run with:

 python annotations/ classifications/classifications.csv

The classifications are then imported into the vandam-data subdataset using ChildProject:

 $ tail -n 1 vandam-data/metadata/annotations.csv 
zoo,BN32_010007.mp3,0,0,50464512,BN32_010007.csv,csv,,BN32_010007_0_50464512.csv,2021-07-19 11:14:58,,0.0.1

In case several users have classified the same chunks, the majority choice is retained. You can have a look at the source of the script to see how that works - or to adapt it to your needs!

Other strategies can be considered; in previous work, Semenzin et al. (2020) have reconstructed the original segments by combining the classifications of the 500 ms chunks, while Cychosz et al. (2021) used the classifications of the individual chunks without reconstruction.

Comparing Zooniverse annotations with other annotations

Once the annotations have been imported into the original dataset, you can use all the functionalities of the ChildProject package e.g. for reliability estimations.

For instance, let's day we'd like to test the ability of the VTC to distinguish children from adults, based on the classifications retrieved with Zooniverse.

The compare does just that (look at the source and try it by yourself!):

 python annotations/

Which will output:

Comparing the VTC and Zooniverse classifications

Going big

This example only contains around a hundred subjects extracted from a sole recording. Real-life projects usually involve much more data - typically tens of thousands of subjects. In order to go big, we advise you of the following:

  • Ask Zooniverse for increased subjects quota.
  • If you are using a version control system such as git/DataLad, you may not want to commit the audio chunks. This can be avoided with appropriate rules in a .gitignore file. Versioning too many files within one repository may cripple it and render operations much slower. Also, provided the metadata for the selected chunks and the original recordings are properly stored and backed-up, the audio chunks can be extracted again at any later time if necessary.
  • Some operations such as sampling or extracting chunks may be demanding for large datasets. We recommend performing this step on a cluster using several CPU cores. The ChildProject provides a --threads option for parallel processing.
Title Launching a campaign of annotations on Zooniverse with ChildProject
Authors Gautheron,Lucas;Laboratoire de Sciences Cognitives et Psycholinguistique;ORCID:0000-0002-3776-3373
Description Step-by-step tutorial to launch a campaign of annotations on Zooniverse based on daylong recordings managed with ChildProject.
License MIT License (
References Managing, storing, and sharing long-form recordings and their annotations [doi:10.31234/] (IsSupplementTo)
Keywords daylong recordings
speech data management
annotation campaigns
Resource Type Software