Re: [AUDITORY] Environmental sound database for memory studies

Subject: Re: [AUDITORY] Environmental sound database for memory studies

From: Eduardo Fonseca <eduardo.fonseca@xxxxxxx>

Date: Sat, 8 Aug 2020 16:52:14 +0200

Arc-authentication-results: i=1; mx.google.com; spf=pass (google.com: domain of owner-auditory@xxxxxxxxxxxxxxx designates 132.206.27.101 as permitted sender) smtp.mailfrom=owner-auditory@xxxxxxxxxxxxxxx; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=upf.edu

Arc-message-signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-archive:list-owner:list-subscribe:list-unsubscribe:list-help :precedence:in-reply-to:to:comments:subject:from:sender:reply-to :date:message-id:references:mime-version:approved-by; bh=D6queGWlaVKqVZE/JYIK+fmQ9u0jXfbhuqs5Fseji4o=; b=bTDtDWphSupElExM0QjsUDtsl2M7XbMNgYJzUiv9j0RogN6Lm0fMEgqqrb2m+duLh5 +p9vERlV4BRmNIhp1ipbsCI1EW6tIkvzevJuZxclCCPI55CGQWxxPX516EHNgt4FU/wa kUe+gF+NAso30tHyzPCEJJ3Jke3zplf2EwSrsOIRzhgwabLgvqVFmtu8yKeCQyWrvIA0 cYZyLzjvhBNy/ir4aXe2fs1WpdoDAxIG8uZtFZlcLmg4nwsz+mMh5JPt0yC7uG34cPJm R/6RDl2TTY8TP12o/S2kPkhRWZKtNDjqfSKiPHfDlCHb+vOH6KVxzyva3ttAJgou8K8Y Jgzw==

Arc-seal: i=1; a=rsa-sha256; t=1596946912; cv=none; d=google.com; s=arc-20160816; b=jKHXEjInIpBWPMRrug8W3t/r0CgzEsPlecJy6+/bJ12gfgsmdadq2NtcJqVNME+MTK TFhzn981JgimhvP+BRNPRnKB0BcvQKfwEmrYARrBjUHo+Fzu4Q6ef/oX1RDjdiZrGPRS blln3juCQ34X03flyvYZeI4ursYI6Y+wpIodevmN2yjC7MSYcDp6iQuce2cs/paSGwu7 tbPWrdCoS9n/lZkJVbq9v2mKoIkc0OiWZPwYDe4J75tE++eXEE1g4d5YTNHKehwTovoA q6jJjXNwN7hNP7grBpzwk3S7hAo6OvsLV6yEd3jXhMI37hQizGByQkHI8VBLq9p1DsDv wdjA==

Authentication-results: mx.google.com; spf=pass (google.com: domain of owner-auditory@xxxxxxxxxxxxxxx designates 132.206.27.101 as permitted sender) smtp.mailfrom=owner-auditory@xxxxxxxxxxxxxxx; dmarc=fail (p=NONE sp=NONE dis=NONE) header.from=upf.edu

Comments: To: Kelly Jakubowski <kellyjakubowski@xxxxxxxxx>

Delivered-to: dan.ellis@xxxxxxxxx

In-reply-to: <CAH91fUOO2wuOARetS=zP4=bVJ9bm0YaE9RKs2BYkt9KL2DVwyA@mail.gmail.com>

List-archive: <https://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>

List-help: <https://lists.mcgill.ca/scripts/wa.exe?LIST=AUDITORY>, <mailto:LISTSERV@LISTS.MCGILL.CA?body=INFO%20AUDITORY>

List-owner: <mailto:AUDITORY-request@LISTS.MCGILL.CA>

List-subscribe: <mailto:AUDITORY-subscribe-request@LISTS.MCGILL.CA>

List-unsubscribe: <mailto:AUDITORY-unsubscribe-request@LISTS.MCGILL.CA>

References: <CA+Qp90rhQiDTDRdn8Vu9voV0j3+VsXdG8cD5corUSiH1_7f57A@mail.gmail.com> <CAH91fUOO2wuOARetS=zP4=bVJ9bm0YaE9RKs2BYkt9KL2DVwyA@mail.gmail.com>

Reply-to: Eduardo Fonseca <eduardo.fonseca@xxxxxxx>

Sender: AUDITORY - Research in Auditory Perception <AUDITORY@xxxxxxxxxxxxxxx>

Hi Nadia, all

Someone mentioned Freesound as an option. We've been putting efforts in curating audio datasets from Freesound. The last dataset we released is FSDKaggle2019, containing 29,266 audio files annotated with 80 labels of the AudioSet Ontology. It has two sources of data:

a manually labeled portion taken from Freesound (you may probably be more interested in this portion)
a portion with noisy labels taken from soundtracks from Flickr videos (the dataset was originally used to run a machine learning competition to study learning from noisy labels)

You can download FSDKaggle2019 and inspect detailed information here: https://zenodo.org/record/3612637#.Xy64VxMzZTY

Also, if you are interested, next month we'll release another dataset of 100h+ hours of manually labeled audio encompassing 200 classes of every day sounds (a superset of the manually labeled portion of FSDKaggle2019).

Another option is the AudioSet dataset, with a massive amount of data and classes.

best,

Eduardo

Eduardo Fonseca
Music Technology Group
Universitat Pompeu Fabra

On Thu, 6 Aug 2020 at 10:24, Kelly Jakubowski <kellyjakubowski@xxxxxxxxx> wrote:

Just to add briefly to Valeriy's quite comprehensive list: We ran a study a little while back, including some participants from this list, to collect emotion ratings from a set of environmental sounds for a project where we needed particularly long sounds (20 seconds), as other work has typically used stimuli around 1-6 seconds. We sourced the sounds from https://www.soundeffectsplus.com/ and I've shared the data on OSF here: https://osf.io/dhk8z/.

Best wishes,
Kelly
---
Dr. Kelly Jakubowski
Assistant Professor (Research)
Leverhulme Early Career Fellow
Department of Music
Durham University
01913341546
musicscience.net

On Wed, Aug 5, 2020 at 5:45 AM Nadia Paraskevoudi <nadiaparask@xxxxxxxxx> wrote:
Dear list,

I am looking for a database of environmental sounds to be used for an auditory memory experiment. Do you have any recommendations?

Thanks in advance!

Nadia Paraskevoudi, PhD student
Brainlab - Cognitive Neuroscience Research Group
Department of Clinical Psychology and Psychobiology
Institute of Neuroscience
University of Barcelona
P. Vall d'Hebron 171, 08035 Barcelona, Spain
Tel:+34 933 125 043
nadia.paraskevoudi@xxxxxx
http://www.ub.edu/brainlab