Dataset Collection

[ynop]

Just a list to collect available audio datasets.

Collections

• http://www.audiocontentanalysis.org/data-sets/
• https://github.com/JRMeyer/open-speech-corpora
• http://www.openslr.org/resources.php
• https://github.com/jim-schwoebel/voice_datasets

Varia:

• AVA Speech (http://research.google.com/ava/download.html#ava_speech_download)

Environmental Sounds:

• https://github.com/karoldvl/ESC-50
• http://urbansounddataset.weebly.com/urbansound8k.html
• https://data.vision.ee.ethz.ch/cvl/ae_dataset/
• Freesound Dataset (https://datasets.freesound.org)
• MIVIA (http://mivia.unisa.it/datasets/audio-analysis/mivia-audio-events/)
• DESED (https://github.com/turpaultn/DESED)

Music:

• http://mirg.city.ac.uk/codeapps/the-magnatagatune-dataset
• https://github.com/mdeff/fma
• https://sigsep.github.io/musdb
• http://medleydb.weebly.com/
• http://mtg.upf.edu/ismir2004/contest/tempoContest/node5.html
• DALI: https://github.com/gabolsgabs/DALI

Speech:

• TIMIT: https://github.com/philipperemy/timit
• TED-lium: http://www-lium.univ-lemans.fr/en/content/ted-lium-corpus
• TUDA: https://www.inf.uni-hamburg.de/en/inst/ab/lt/resources/data/acoustic-models.html
• LibriSpeech: http://www.openslr.org/12/
• AMI Corpus: http://groups.inf.ed.ac.uk/ami/corpus/
• SIWIS: http://www.unige.ch/lettres/linguistique/research/latl/siwis/
• Buckeye: https://buckeyecorpus.osu.edu/
• CSTR VCTK: http://homepages.inf.ed.ac.uk/jyamagis/page3/page58/page58.html
• Mozilla Common Voice: https://voice.mozilla.org/en/data
• Free Spoken Digit Set: https://github.com/Jakobovski/free-spoken-digit-dataset
• Vystadial 2013: https://lindat.mff.cuni.cz/repository/xmlui/handle/11858/00-097C-0000-0023-4671-4
• https://tatoeba.org/eng
• The Spoken Wikipedia Corpora (https://nats.gitlab.io/swc/)
• QUESST-2014 (http://speech.fit.vutbr.cz/software/quesst-2014-multilingual-database-query-by-example-keyword-spotting)
• https://github.com/nicolaspanel/TrainingSpeech
• SWS2013 (http://speech.fit.vutbr.cz/software/sws-2013-multilingual-database-query-by-example-keyword-spotting)
• CSS10 (https://github.com/Kyubyong/css10)
• CMU Wilderness(https://github.com/festvox/datasets-CMU_Wilderness)
• M-AILABS (http://www.m-ailabs.bayern/en/the-mailabs-speech-dataset/)
• AudioMNIST (https://github.com/soerenab/AudioMNIST)
• COSINE (http://melodi.ee.washington.edu/cosine/)
• Fluent Speech Commands (http://www.fluent.ai/research/fluent-speech-commands/)
• Scotus Speech (https://github.com/noajshu/scotus-speech)
• Living Audio Dataset (https://github.com/Idlak/Living-Audio-Dataset)
• VOiCES (https://voices18.github.io/downloads/)
• EUSpeech (https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/GKABNU)
• LibriVoxDeEn (https://www.cl.uni-heidelberg.de/statnlpgroup/librivoxdeen/#beilharz19)

[emzee831]

Hello, I am new to the gitlab collaborating world and coding. Very eager to help out and gain some experience. How can I help out with this issue?

[ynop]

Hi, if you like you can implement a reader for another dataset/corpus from the list above or any other one you know.

You can find examples of existing implementations in https://github.com/ynop/audiomate/tree/master/audiomate/corpus/io.

There is some documentation on adding new datasets as well:
https://audiomate.readthedocs.io/en/latest/documentation/new_dataset_format.html

This one (http://www.openslr.org/1/) for example looks like a simple one to start.
Let me know if you need any help.

If you start, please open a new issue for further discussions.