Group

Acoustic Monitoring / Feed

Bioacoustic monitoring is one of our biggest and most active groups, with members collecting and analysing acoustic data from every type of wildlife, from birds and bats to big cats, and even reptiles!

discussion

Help with AudioMoth configuration for bat recording

Hi! I am trying out some different settings with the AudioMoth configuration app to see the most suitable setup for recording bat species. After the device is configurated, I...

1 0

Koen,

Hi - the config for the audiomoth looks fine (you could put a high-pass filter on to cover frequencies 12-96 kHz only). The scheduler is not set (you have 0 files per day), but presume this is just the screenshot.

The issue you have, though, is to do with your loudspeaker - presuming it is just a standard one. Standard music/computer speakers are tuned to human range of hearing and don't emit sound above around 20 kHz.  They won't emit ultrasound to 100 kHz.  For that you need specialised equipment.  If you want to generate some ultrasound yourself though, just rub your fingers together, or jangle some keys near the Audiomoth!

Hope that's of some use, Carlos

See full post
event

BirdCLEF 2022 Kaggle Challenge

The Cornell Lab of Ornithology
Join the BirdCLEF 2022 Kaggle challenge, where you’ll use your machine learning skills to identify bird species by sound. You'll develop a model that can process continuous audio data and then acoustically recognize the...

0
See full post
discussion

Acoustic monitoring - sharks

Would anyone know about the recommended number of months to conduct acoustic monitoring of sharks. This is part of a small grant project. Nine months are being proposed to...

0
See full post
article

Introducing the WILDLABS On the Edge Fellows

WILDLABS Team
We're proud to introduce the first WILDLABS On the Edge Fellows for 2022, Loretta Schindlerova and Meredith Palmer! Working alongside expert Edge Impulse mentors, these two fellows will use embedded machine learning to...

0
See full post
discussion

Using rechargeable lithium-ion batteries for AudioMoths and/or Swift's

Hi all, I'm wondering if anyone has experience using rechargeable lithium-ion batteries for passive acoustic devices, specifically Audiomoths (AA batteries) and Cornell...

12 0

Actually it's interesting. I think it might be a good opportunity to think about how we can transition to rechargeable batteries in the field. Shipping them is possible if you follow postal rules and pay for hazardous material charges. But it needs to be planned in advance. Anyways, might be a good conversation to have in the community. But there also needs to be other foundational work like understanding how they're used, getting chargers in the field, and how to interface rechargeables to standard equipment.

I've taken rechargeable batteries on planes no problem - NiMH can go in checked luggage. I use NiMH D batteries for the Swift's, because they take more and so it still lasts long. The Audiomoth is harder, because it's just the 3 AAs and even with lithium-ion you only get ~1 month. The drop off to NiMH is like a week, from my (minimal) trial-and-erroring.

We've been using 18650 lithium ion protected cells in our BAR recorders since 2013. The protected cells are great because the protection circuitry kicks in to disconnect the battery and save it from being dicharged too much. It also protects against short circuit.

So if you did want to give 18650 batteries a try then I'd recommend a Keystone 1043P (P for polarised so you can only fit the battery one way)  battery holder and KeepPower protected 18650. There are a few other brands too that fit that battery holder ok but that's a common consistent brand.

Eizfan and XTAR make good chargers.

Most of our customers use them in the same continent and drive or road freight their recorders to site and so don' thave any problems shipping them.

We've heard of some problems on planes taking them as hand luggage and it's just luck of the draw if the particular person you encounter in the baggage area knows what their lithium ion battery policy is. I've usually heard of I think attempts to get this claried at the airline level has lead to a run around too.

Some people get to check them in if the batteries are fitted into the battery holders because they consider it built in like a phone or laptop. The spare batteries often go on as hand luggage. We have plastic battery cases to hold the cells for travel.

If you're sending it overseas we find that DHL, UPS and Fdex seem to take about a month to get the paperwork sorted out to give you permission on your account to ship lithium ion batteries with your recorders. Once you have permission though you can do it whenever you need to in the future for subsequenct studies. Look up the IATA regulations for packing lithium ion batteries with your equipment. You'll need to know this so you can tell DHL etc what permission you need and what you typically do. We use UN3481 packing instruction 966 section II becasue our cells are <20Wh and we can ship 5kg of batteries in a carton along with our recorders.

https://www.iata.org/en/publications/newsletters/iata-knowledge-hub/what-to-know-about-how-to-ship-lithium-batteries/

Mark Calder

Frontier Labs www.frontierlabs.com.au

See full post
discussion

Special session on "Open-Source and Free Tools for Bioacoustics" at the ASA Meting in Denver

The 182nd meeting of the Acoustical Society of America will be held in Denver, CO, from 23 to 27 May 2022. During this meeting there will be a special session on ...

2 0

Do you know if there are plans for remote attendance?

See full post
event

Bioacoustics Webinar: Primates You May Have Missed

Wildlife Acoustics
Join Wildlife Acoustics on January 20th, 2022, at 3-5pm GMT for their free webinar, Primates You May Have Missed. At this event, participants will learn how bioacoustics can be used for primate research. 

0
See full post
article

GroupGets: AudioMoth Underwater Case

Open Acoustic Devices
Don't miss your chance to order the AudioMoth Underwater Case through GroupGets! This campaign is for a waterproof case designed specifically for AudioMoth version 1.2.0 (without the 3.5mm socket), and for use at depths...

0
See full post
article

New Paper: HydroMoth Testing

Remote Sensing in Ecology & Conservation
Open Acoustic Devices have tested their new HydroMoth, an affordable marine acoustic tool that could open up new avenues for marine conservationists without budget for expensive hydrophones. Read the newly published...

0
See full post
article

Deep Learning for Marine Ecology and Conservation

arXiv (Journal)
This article provides a review of deep learning (predominantly ML) used in marine ecology and considerations for its future directions in conservation. In plain language, the authors provide a methodology for training...

0
See full post
discussion

batch/automated cloud processing

I saw another thread asking about getting BirdNet (https://github.com/kahst/BirdNET) running on a Windoze machine, and thought I would expand on that question and ask what people...

14 0

Hi Antoine, did you know about our bioacoustic recorders? https://frontierlabs.com.au/ . They have the best sound recording quality and built in GPS. You can have a listen to Australia here https://data.acousticobservatory.org/projects/1 . The Australian Acoustic Observatory uses 400 of our Solar BARs across Australia and the recordings are published for anyone to use.

Unfortunately there isn't software to do an easy bird census. Do you  need to know every bird you can hear or just the presence of one species?

Best regards, Mark

Wanted to chime with another option. My lab develops OpenSoundscape, a Python software package for training machine learning models for bioacoustic detection, especially convolutional neural networks. Check it out at http://opensoundscape.org/

 

I see that the list of bioacoustic softwares was mentioned earlier in the thread (thanks Carly!) - I'm always on the lookout for new softwares to add to that, so drop me a line if you see a package that is currently maintained and relevant to bioacoustics but is not included on the list. 
(Here's the link: https://github.com/rhine3/audiomoth-guide/blob/master/resources/analysis-software.md)

Hi all,

For everyone's reference but also a new one for your list Tessa (just checked and it isn't there yet) - 

Fuentes, M., Salamon, J., Zinemanas, P., Rocamora, M., Paja, G., Román, I. R., ... & Bello, J. P. (2021). Soundata: A Python library for reproducible use of audio datasets. arXiv preprint arXiv:2109.12690. https://arxiv.org/abs/2109.12690. 

 

See full post
article

MBARI shares trove of marine acoustic data

Monterey Bay Aquarium Research Institute
The Monterey Bay Aquarium Research Institute has an extensive acoustic archive from underwater recordings and has made their valuable dataset available to researchers around the world via an open data registry on the...

0
See full post
funding

Opportunity: μMoth v1.1 Group buy

Group Gets
GroupGets has a Group buy campaign for a μMoth v1.1(Round 2), that runs till  Sat, 06 Nov 2021 12:55:00 PDT. The μMoth v1.1 goes for a Single-unit price of  $99.99 + shipping.To join this buy, kindly visit this page

0
See full post
discussion

What value would you place on regular status updates from your acoustic recorder?

Frontier Labs is currently weighing up the possibility of including a 'heartbeat' function for our next iteration of our...

5 0

Hi Astaras, I'm Mark Calder one of the engineers at Frontier Labs. Yes we think that the energy used in sending a message by satellite once per week will get quickly replaced with the solar panel. We plan to send a short message with battery capacity remaining and card space remaining once per week typically. Probably also that sound was recorded in the last recording period. That is the microphone signal output was not flat fine. That can happen if the mic is broken off for example.

Obviously if you don't get a message from one of your recorders at the set time then that's cause for concern. It can take 4 hours to aquire a satellite and send the message though so you'd probably wait until the next day to make that judgement. The battery in a Solar BAR lasts for about a month with 24/7 recording if there is no more solar so it should be able to send out messages during long rain periods.

We'll be starting this project later this year. If you can think of other info to have in the messages please let me know. My email is [email protected] if that's more convenient.

Hi Nicole--

We've already chatted about this function, and I think the answer is "quite a bit" for a lot of deployments. The more accurately the proposed function tracked microphone status the better, though even something to the effect of "I'm still getting a signal from this microphone" even if there was no more nuanced quality information than a yes/no would still be better than what much of the field is doing right now. 

I think the question is how much this would add to the cost of the unit + how much whatever data service you're using would add to the ongoing costs but I'd imagine that Frontier are still figuring out those specifics.

See full post
discussion

Arbimon terms of service

I'm currently working on trying to get my lab group to start using Arbimon, since it has some pretty cool functionality. However, in the Arbimon terms of service, I ran across...

2 0

Hi David!

It sounds like Rainforest Connection and its affiliates reserve the right to use your acoustic data and its labels to train and/or test sound identification algorithms.

In my opinion, Rainforest Connection is unlikely to "scoop" the direct ecological or conservation research outputs of your data, which could be what your collaborators would be most concerned about. 

However, if one of your research goals is methods development (e.g. publishing trained sound classifiers or experimenting with different machine learning approaches), you may want to directly ask Rainforest Connection for clarification on this part of their terms of service.

Sidenote: one the most significant bottlenecks for developing accurate machine learning classifiers is a lack of labeled data. The bioacoustics community would certainly benefit from the open publication of more labeled data!

Tessa

Tessa--

Thanks for chiming in. I agree with a lot of what you're saying, both about what RFCX is likely to actually be doing and about the need for more open labeled datasets. However, unfortunately, a lot of times the academic incentives of individual research groups don't align well with the needs of the field more broadly.

This honestly might be a good topic for a broader discussion: how do we balance the ideals of a strongly FOSS-influenced conservation tech field with the incentives of academia, which include both publishing and (in the case of some universities including mine) monetizing data and research.

See full post
event

How do I analyze large acoustic datasets using PAMGuard?

Jamie Macaulay
The recording is now available from our Tech Tutors episode with Jamie Macaulay, who answered the question, "How do I analyze large acoustic datasets using PAMGuard?" In this episode, Jamie taught us how to use the free...

0
See full post
event

Bioacoustics Webinar: Creatures of the Night

Wildlife Acoustics
Join Wildlife Acoustics on October 7th at 10 AM ET/ 3 PM BST for their bioacoustics webinar, Creatures of the Night: How Sound Shines a Light on Species Active at Night. At this event, participants will learn about how...

0
See full post
discussion

Incidental recording of people (in an IRB context)

I'm currently going through an IRB submission for a survey I intend to give landowners as part of an acoustic sensor deployment in upcoming months, and I'm being asked...

2 0

I haven't needed to go through an IRB for acoustic collections yet, but I have had landowners ask this.  My response has been that we filter out incidental conversations with ML automatically.  For an IRB, you will probably need to be specific about how and when you do this.  The responses I have had have been very mixed and any objections are based on incorrect assumptions and knowledge....or a flat out "no way" before even hearing how it works - usually after hearing the words "continuous recording" on their property.  Others are happy and excited to see what is going on.  It seems to be an educational task and it isn't easy amidst the current privacy debates, not to mention political climate.

Please keep the board updated.  I'm sure your IRB experience with acoustic monitoring deployments will be valuable to others here.

We are filming with the Australian Acoustic Observatory soon with one of the Solar BARs we created (it's continuously recording for four years!) and were wondering if recording our people chat would constitute audio vandalism? lol

See full post
discussion

Deep learning module for PAMGuard

Hi all We've recently been developing a new deep learning module for PAMGuard which will be released soon. For those of you unfamiliar with PAMGuard, it's an open...

2 0

Thanks heaps for the info @jamie_mac ...looks really cool and the blog is amazing too!

I dare say you might see quite a few excited responses from the community!

All the best,

Rob

Thanks Rob! Hopefully it turns out to be a useful tool.

 
Just to reiterate for folk, there are links to tutorials and a beta version in the blog post.

See full post
event

Webinar: Bioacoustics and Engagement

Wildlife Acoustics
Join Wildlife Acoustics for their webinar, Super Hear-os: Engaging Everyday People Through the Science of Bioacoustics, this Thursday, July 22nd at 10 AM ET / 3 PM BST! At this event, four presenters will discuss how...

0
See full post
event

WILDLABS Tech Tutors: Season 3

WILDLABS Team
The WILDLABS Tech Tutors are back for all new season of in-depth walkthroughs, deep discussions on effective, impactful, and inclusive conservation technology project strategies, and - of course- even more answers to...

0
See full post
article

GroupGets: μMoth Now Available

Open Acoustic Devices
The new μMoth is now available on GroupGets! These units are going quickly, so join the order quickly to be among the first to try the world's smallest full-spectrum acoustic development board! Join the GroupGets order...

0
See full post
discussion

Tools similar to BirdNet for analyzing avian recordings?

Hello everyone,  I am a new member, and this is my first post.  I am an amateur conservationist and birder, and have a love of birding (though not quite the ear to go...

5 0

Almost a year since the original post, but just commenting that I've been having some pretty good success with BirdNET which has been working fine in the past few weeks I've been using it.

Dear Thomas I'm really interested by your experience, how did you use birdnet?

I would like to monitor a area for a certain amount of time with birdnet automatically identifying the bird around, is it possible to do that? Or did you record a sound bird for a while, then transfer the audio file into bird net and manually identify each bird sound?

All the best, Antoine

Tessa Rhinehart has developed a great compilation of bioacoustic analysis platforms, software, etc. with basic functionality on each. It's openly-available on Github -https://github.com/rhine3/audiomoth-guide/blob/master/resources/analysis-software.md. 

I personally use RavenPro (from the Cornell Lab of Ornithology), Kaleidoscope (from Wildlife Acoustics), and Arbimon (from Rainforest Connection) in different capacities. I've also heard good things about PAMguard (developed firstly in the marine mammal community) and Sonic Visualizer.

See full post
discussion

BirdNET on local Windows machine?

Hi there, did anyone get the https://github.com/kahst/BirdNET running on a Windows machine? We get good results online and would love to run it locally. Greetings from Vienna...

3 0

Dear Robin,

 

So what is your workflow regarding the utilisation of birdnet to analyse sound bird?

 

All the best,

Dear Antoine,
collect sound data (.wav) with the audiorecorder (audiomoth) of your choice.
Point the script https://github.com/kahst/BirdNET-Lite to the location of
your data storage (MicroSD) attached to the Linux machine, wait one day and see the magic happen ;-)

Greetings,
Robin

See full post