LightNews
ricard.bsky.social
Ricard Marxer
@ricard.bsky.social
440 followers 190 following 13 posts
ml, audio, cv, nlp, speech, bioacoustics // Assoc. Prof. at Université de Toulon, researcher at LIS CNRS UMR 7020, director of http://www.master-mir.eu in marine robotics and AI
www.master-mir.eu
Posts Media Videos Starter Packs
Reposted by Ricard Marxer
arikkershenbaum.bsky.social
Arik Kershenbaum @arikkershenbaum.bsky.social · Jul 8
New paper! Using AI to track the frequency of vocalisations across a (very) wide range of species
www.ricardmarxer.com/f0-bioacoust...
In wolves, dolphins, etc, the up-down of frequency of their calls is where all the information is. Neural nets can do this automatically!
‪
1 2 3
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Jun 20
‪@brbbfreitas.bsky.social‬ ‪@frantsjensen.bsky.social‬ @arikkershenbaum.bsky.social‬ @malavika-m.bsky.social‬ @hollyhowls.bsky.social‬
1 1
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Jun 20
Robert C. Liu, Malavika Madhavan, Andrew Markham, Marie A. Roch, Holly Root-Gutteridge, Martin Šálek, Grace Smith-Vidaurre, Ariana Strandburg-Peshkin, Megan R. Warren, Matthew Wijers
1 1
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Jun 20
🔗 Project page: www.ricardmarxer.com/f0-bioacoust...
📄 Paper: doi.org/10.1080/0952...

Fantastic work with Paul Best and Marcelo Araya-Salas, Axel G. Ekström, Bárbara Freitas, Frants H. Jensen, Arik Kershenbaum, Adriano R. Lameira, Kenna D. S. Lehmann, Pavel Linhart
Benchmarking Neural F0 Estimation for Bioacoustics
Bioacoustic fundamental frequency estimation: a cross-species dataset and deep learning baseline
www.ricardmarxer.com
1 1
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Jun 20
We provide:
🔧 A tool with pretrained models to estimate F0 on new data
📊 An annotated dataset of 250k+ vocalisations
1 2
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Jun 20
Key findings:
✅ Neural network models show strong potential
✅ Supervised models generalise to unseen species
✅ Self-supervised models can perform on par with supervised ones in some cases
✅ As expected, performance correlates with F0 saliency
1 1
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Jun 20
The diversity of bioacoustic signals—in frequency range, harmonicity, SNR, etc.—makes F0 estimation particularly challenging.

We study how well #MIR F0 estimation transfers to non-human vocalisations, benchmarking SOTA models across 14 taxa:
1 1
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Jun 20
📢 New paper!

F0 estimation is a classic task in the Music Information Retrieval (#MIR) community. Yet, it's much less studied in bioacoustics—despite its frequent use to describe animal sounds.

Images from doi.org/10.1038/srep... doi.org/10.1098/rspb... doi.org/10.3389/fmar...
2 7 16
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Jun 7
Our study on #orcas vocal complexity!

doi.org/10.1016/j.ec...

We analyse #bioacoustics recordings spanning 5 years of wild orcas. DL classification and subsequent complexity measures are contrasted to pod sizes, to approach the social complexity hypothesis in #cetaceans

Good collab Best et al!
1
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · May 19
Happy to see this work out there. Good job @charlylamothe.bsky.social et al. it was really nice working with you all!
charlylamothe.bsky.social
Charly Lamothe @charlylamothe.bsky.social · May 19
I'm pleased to announce the publication of my PhD works, a large annotated dataset of vocalizations by common marmosets. With over 800,000 recordings and 253 hours of data, this is the largest annotated dataset. Check it out and let me know what you think!
www.nature.com/articles/s41...
A large annotated dataset of vocalizations by common marmosets - Scientific Data
Scientific Data - A large annotated dataset of vocalizations by common marmosets
www.nature.com
1
Reposted by Ricard Marxer
danstowell.mastodon.social.ap.brid.gy
Dan Stowell @danstowell.mastodon.social.ap.brid.gy · Mar 21
Come and work with us! Be a Postdoctoral Fellow in AI for Ultrasonic Bioacoustic Monitoring! All the details: https://www.naturalis.nl/en/about-us/job-opportunities/postdoctoral-fellow-in-ai-for-ultrasonic-bioacoustic-monitoring #academicjobs #postdoc #pdra #netherlands #biodiversity #ai […]
Original post on mastodon.social
mastodon.social
1 18 18
Reposted by Ricard Marxer
bencretois.bsky.social
Benjamin Cretois @bencretois.bsky.social · Jan 13
Hi! 👋
I am Benjamin and I am a researcher at the Norwegian Institute for Nature Research (NINA).
I work at the intersection of #statistics, #AI methodologies (deep learning really) and #biodiversity monitoring - more specifically #bioacoustics 🦜🔊
2 5
Reposted by Ricard Marxer
danstowell.bsky.social
danstowell.bsky.social @danstowell.bsky.social · Dec 18
Academic Job! Come and work with us as "Assistant Professor, Data Science & AI" tiu.nu/22594 - friendly department, lots of stimulating colleagues, Dutch work-life balance... oh and the #AI & #datascience #research :) #jobs #academicjob #netherlands
Job opening: Assistant Professor Data Science & Artificial Intelligence 1.0 FTE (22594)
tiu.nu
2 7
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Nov 13
Today Santiago Cuervo and I are presenting our study on "Scaling Properties of Speech Language Models" at #EMNLP2024. More details soon. Can't wait? come chat with us at Poster session D.
4
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Oct 17
Check out our latest survey on the use of #machinelearning for #bioacoustics ! We aimed to make it interesting to both biologists and computer scientists.

I had a great time working with this team.

doi.org/10.1111/brv....
Automatic detection for bioacoustic research: a practical guide from and for biologists and computer scientists
Recent years have seen a dramatic rise in the use of passive acoustic monitoring (PAM) for biological and ecological applications, and a corresponding increase in the volume of data generated. Howeve...
doi.org
1 1
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Mar 1
So happy to finally be presenting this work with Clementin Boittiaux et al!

SUCRe uses SfM/MVS dense reconstruction to recover colors from underwater images.

🔗 clementinboittiaux.github.io/sucre/

To appear at #3DV2024 in March.

#underwater #deepsea #ai #machinelearning #computervision
3
ricard.bsky.social
Ricard Marxer @ricard.bsky.social · Oct 27
Become a 4th wave student of our #ErasmusMundus #MSc in Marine and maritime Intelligent Robotics (MIR)

An international program combining #AI and #Robotics for #Ocean applications 🧠🤖🌊

Applications open tomorrow October 28th 2023

More here: www.master-mir.eu/emjmd-mir-ne...
3