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

Prior work on Bird Flock identification

Hi there! We have a project underway called "Identi-Flock" which is an ambitious attempt to port our individual pollinator identification software, www.withymbe.info...

5 0

Hi Andrew,

Dan here—I'm one of the authors of the TinyML book! I love your Withymbe project; I've previously done work involving embedded systems and insects, and it's interesting to hear about your plans for bird flocks.

As long as you have sufficient data, you should be able to identify different bird sounds and discern them from background noise. The TinyML book has a chapter that introduces the underlying techniques, and I'd also recommend taking a look at www.edgeimpulse.com - we've built a set of tools designed to make it easy to train these types of models.

We actually recently published a tutorial on Wildlabs about this very concept:

https://www.wildlabs.net/resources/case-studies/tutorial-train-tinyml-model-can-recognize-sounds-using-only-23-kb-ram

I'm always excited to learn about new applications; feel free to reach out if there's any way we can help. I'm [email protected]

Warmly,
Dan

Just guessing but I don't think it will make much of a difference, individual or flock.  The spectrogramme will look much the same, and I think that is used as the input vector to the CNN.  If so then I would expect the model will be quite tolerant of flock size.  Just spitballing here though.

Hi Harold!

 

Great to know you are in the domain. To be honest my analysis so far indicates that when conducting a DSP approach on the spectrum, smoothing via convolution becomes an issue? Basically, the raw spectrum is too jagged to match, so one convolves it to smooth it, but then one just gets a generic "noise"-shaped spectrum. I also have variances in sampled spectra from the same source recording? I am using an fs=44100 and a spectrum 0 - 64kHz initially, or though I tried to filter from 100 - 9k with little success?

My design outline is: I need to identify the presence of a flock of a certain species of avians, I need to know when the flock is not present, and I need to distinguish the presence of other flocks of birds, not to identify them, but they are sometimes similar in size and possibly, therefore, call range? A sort of "We - Not We" approach?

I am comparing the gestalt sound, not individual calls?

Plus: I am using a Rapsberry Pi for the Fog Node currently, but see that I can use my Arduino Uno for TinyML from the examples which use a Nano? I am interested in the power-saving, but need a robust microphone rig, which I currently get via usb?

I will checkout your tutorial, many thanks!

Tally ho!

Andrew.

See full post
article

Enter the Zooniverse: Try Citizen Science for Yourself!

Ellie Warren
Trapped inside during the COVID-19 quarantine and looking to engage with conservation science without leaving your desk? Citizen science projects like those on Zooniverse offer a great opportunity to impact scientific...

0
See full post
discussion

Success recording bees using AudioMoth

I would just like to put on record that we have successfully tested an AudioMoth recording "bug" in a local garden inside a plastic bag whereby it recorded the buzz of...

1 0

How can we learn more about your BEESWAX7 buzz identification and counting program, and discuss working together?

See full post
article

#Tech4Wildlife 2020 Photo Challenge In Review

WILDLABS Team
2020 marked our fifth year holding our annual #Tech4Wildlife Photo Challenge, and our community made it a milestone to remember. Conservationists took to Twitter last week to share their best high-tech snapshots from...

0
See full post
article

Listening to Nature: The Emerging Field of Bioacoustics

Adam Welz
Researchers are increasingly placing microphones in forests and other ecosystems to monitor birds, insects, frogs, and other animals. As the technology advances and becomes less costly, proponents argue, bioacoustics is...

0
See full post
event

Workshop: Using Bioacoustics for Field Survey

CIEEM
CIEEM is hosting a one-day workshop in bioacoustic field survey methods at a beginning to intermediate level. Attendees will attend hands-on workshops, lectures, and computer sessions giving an overview on using...

0
See full post
event

WILDLABS Virtual Meetup Recording: Acoustic Monitoring

WILDLABS Team
The fourth and final event in Season 3 of the WILDLABS Virtual Meetup Series is now available to watch, along with notes that highlight key takeaways from the talks and discussion. In the meetup, speakers David Watson, ...

0
See full post
event

Acoustic Telemetry Workshop Series

Ocean Tracking Network
Ocean Tracking Network and ideasOTN are hosting an acoustic telemetry workshop in Halifax in February 2020 to assist early career researchers to develop skills and build their toolbox. Topics include deployment, YAPS...

0
See full post
discussion

Machine-readable representation of a sensor duty cycle / recording schedule

My group is currently trying to improve the database we use to store our recordings, and we're trying to figure out how to represent a sensor's duty cycle in a way that is...

1 0

Hi David,

I use cron-like patterns.  One form looks something like: mon,wed-sun 0-23:0-59/10 90

which means every day but Tuesdays, every 10 minutes from the top of the hour, recording for 90 seconds.  There is a similar form for matching by month/day.  There could be yet another form for sunrise/sunset but I haven't implemented that.  Basically a hyphen indicates a range, a comma indicates a list, and the slash is repetition step size.  A full recording specification consists of one or more patterns, the first matching being used (for the recording duration).  Years are not implemented but that could be done also.

Thanks,

-harold

See full post
discussion

Advice on Audiomoth extended power and case solution

I am leading a project to conduct canopy camera trap and audio monitoring for birds and mammals along a conservation corridor at the Jama-Coaque Reserve in coastal Ecuador. We...

3 0

Hi Shawn,

I might be concerned about the Li-ion batteries, if there is any chance of losing a node (damage due to installation, weather, animals, or theft) then the economics won't work out, it being such a large proportion of the cost.  Also it's kind of a pain flying with them even if they're quite light because they have to go in carry-on luggage.

I'm not insensitive to the e-waste issue, but here I think even here alkaline might have the upper hand: many lithium cells contain toxic cobalt (as opposed to manganese) and it's not often clear which type you have.  Alkalines OTOH are relatively innocuous (unless they contain mercury, which they should not in this day and age): carbon anode, steel casing, zinc cathode, zinc oxide (used in baby nappy rash cream), lye, manganese oxide/dioxide (occurs naturally as an ore but you still don't want to eat it).

Thanks,

-harold

Many thanks Harold!

I really appreciate the information on e-waste! What about going with LiFePo4 batteries? They are consider low toxicity and I know they are a type of battery that our chemistry department will take to recycle (they will also take lipo and li-ion). I am trying to figure out how to calculate battery life in the Audiomoth if I were to go with a higher voltage LiFePo4 but less mAh, say 6.6 volt and 2100/3000 mAh. Any suggestions? and of course thoughts on this option? These batteries are cheaper too ($11-15)!

I've worked a lot with lipo and li-ion batteries building drones and other power solutions so I'm not really worried about damaging them or transporting them. It's amazing how many huge lipo's I've had in my carry-on luggage and all they care about is my laptop! My main concern with alkaline is that they are expensive in Ecuador with no where to dispose of for recyling (yet or that I currently know of). Rechargeables make sense because of the long-life and they can be muled back to the university for recyling with the chemistry department (I have no idea what they do with them, they just want them and say they will be recycled).

Shawn

Hi Shawn,

I've only heard good things about LiFePO4.  The only issue I can think of is exactly the one you bring up: the EMF is low enough that a 2S battery is needed.  The Audiomoth uses a linear voltage regulator, so anything in excess of the 3.3V that is actually used is wasted.  This means for endurance calculations you consider only the mAH rating given you have chosen a battery that will output at least 3.3V (plus a bit more for the dropout).  So 2100mAH at 6.6V is just 2100mAH, compared with 5000mAH of your initial Li-ion.

A high-effiency switching regulator may make sense if you're running the Audiomoth at high duty, but even so finding a low power unit is hard.  But since reliability is paramount, I would go with something with the fewest moving parts.

Thanks,

-harold

See full post
discussion

Audio analysis of flocks or swarms

Hi, Here at BEESWAX8 we are working on identifying flocks of avains or swarms of insects by their collective noise. We have noticed already that this is more complex than...

2 0

Hi Andrew,

I'm not a signals guy I'm afraid, but I was thinking perhaps one way might be to estimate the physical size of the flock/swarm by scanning with a microphone array.

BTW do you find it a problem if an individual happens to be too close and drowns out the rest?  In part it's a question of dynamic range but also of sensor placement.

Thanks,

-harold

So far we have been able to disambiguate between the flocks of two different species of small birds? Not easy. Our problem seems to be based around the fact that some or all flocks of small birds cheeping sound pretty similar? Of course the issue is: at what point is asynchronous sound aggregation just noise?

With regards to individual sounds, we are not interested in individuals at the moment so have not tested for it. I suspect however that in the interests of completeness we will be?

More anon,

Andrew.

See full post
discussion

Raspberry Pi Hardware Mono Mic Problem

Hi all, I'm building a passive bioacoustic device for a PhD project and I have come across an issue for which I need advice from someone with more audio tech experience....

5 0

We've had good results with two mono inputs accepted via the stereo 1/4" input jack on the Pisound HAT on a 3B+ and just about to test with the 4. Details and links to Pisound folks in Lithuania here -- http://www.orcasound.net/2018/04/27/orcasounds-new-live-audio-solution-from-hydrophone-to-headphone-with-a-raspberry-pi-computer-and-hls-dash-streaming-software/

Hi Sam,

have you tried the Clippy EM172 ? I was looking at this microphone on Raspberry Pi + PiSound, but it has an electret capsule and should then be powered...

Thank you !

See full post
discussion

Sound Event Recognition - through collaboration

here's our 90-second video explaining the SERVAL sensor, which saw the light thanks to our collaboration with Karol Piczak. The 7th of July we'll organize a...

6 0

@jankees,  

Hi I am working on a system to reduce Human Elephant Conflict and am interested in acoustics of Elephants and the identification of elephants through their acoustic signature to assist with this project, did you get anywhere with the serval sensor for detecting Elephants? 

We are a non profit called the Forgotten Parks Foundation and currently managing the Upemba and Kundelungu National Parks in the DR.Congo and have an immediate requirement to reduce HEC so we would like to assist with the development of such a system. 

Hi Kev,

thank you for contacting us. We are a non-profit, too. I very much welcome collabortion on this system. Shall we continue our discussion through email? Mine is [email protected] 

Looking forward to discussing this further,

Jan Kees

See full post
event

WILDLABS TECH HUB Showcase

WILDLABS Team
Join us at the Tech Hub Showcase event ot hear how our winners are using technology to scale their solutions to the illegal wildlife trade. The event will take place at Digital Catapult, 101 Euston Road, London, on the...

0
See full post
article

Meet the WILDLABS TECH HUB Winners

WILDLABS Team
In February, we released an open call for the WILDLABS TECH HUB, offering 3 months of support for solutions using technolgy to tackle the illegal wildlife trade. We were overwhelmed by an incredible 37 submissions,...

0
See full post
discussion

AudioMoth 2.0 Development

Hi. I'm moving a topic to this thread from the camera trap thread since it's more relevant here.  Discussion with Alasdair for context: "Lastly -...

0
See full post
discussion

AudioMoth user guide

AudioMoth users, I recently wrote this guide to operating and deploying AudioMoths: ...

9 0

That also sounds like a good idea :)

Hi again,

I agree; it's very important that everyone is able to contribute to the guide. Thanks for taking the lead on that, Akiba. To that end, I just made my GitHub repo publicly editable, so please feel free to add to it if you want to use a GitHub repository.

However, a Google Doc would be much easier for the general public to contribute to. If you choose that route instead, please use whatever you want from my repository. :-) 

I personally like to use the guide as a single document instead of a Wiki. It's easier to read through, and since everything is on the same page, you can just CTRL+F to find a specific word or section. Also, it's helpful to be able to download it as a .pdf, or print it out and refer to it in the field. 

Tessa

Hi Tessa. 

Thanks and also thanks for being so generous with your work :)

Regarding the format, how do you all think if we have a single document and a wiki? I could handle the single document if David wanted to do the wiki. One potential issue would be synchronization, but since many wikis keep a change list, perhaps we can schedule periodic document updates based on the wiki changelists. We can also maintain a document changelist in case the changes need to go in the other direction.

Any comments or suggestions?

Akiba

See full post
discussion

Advice for an in-situ Audio-Visual system for recording at a nest site

I'm looking for a cost-effective solution to a remote in-situ audio-visual system for bird acoustics for use in a PhD Project. I need an audio-visual setup suitable...

9 0

Hello Carlos and Sam
That book on "CCTV for Wildlife Monitoring" looks excellent. Although it is somewhat off at a tangent of my actual work this looks way too useful not to have. There are cheaper second-hand copies on Abebooks and eBay and also on sale at NHBS, £20.98 incl. UK P&P (https://www.nhbs.com/cctv-for-wildlife-monitoring-book).

An interesting interview with the author, Susan Young: https://blog.nhbs.com/author-interviews/interview/cctv-for-wildlife-monitoring-an-interview-with-susan-young/

Meanwhile, I have come across a research project by the author of that book using technology that looks very close to what Sam is trying to do: "CCTV Enables the Discovery of New Barbastelle Vocalisations and Activity Patterns Near a Roost - We monitored a barbastelle maternity roost for four months using a portable CCTV system, time synchronised with ultrasound recorders."
See more here: https://www.researchgate.net/project/Using-CCTV-to-link-vocal-repertoire-with-behaviour-in-bats

If the above does not do it, I think you should contact the author and get her advice on what would be a suitable setup for you. That's what I meant in my earlier post about maximising use of previously developed solutions. It will just save you a lot of messing about, especially if you are not a technical expert yourself.

I hope this is of some use. Do post an update when you have worked out what equipment to use.

Good luck,
Joachim Neff

Hello Sam

I have now received the book on "CCTV for Wildlife Monitoring" (see my earlier post for more details) and it is as excellent as it looked online. If you can spare £21 then I would suggest you get a copy yourself.

The book is written for those without much technical background and the case studies are very detailed with equipment lists and setup details. I have not read the entire book but as far as I can see there are no examples described specifically with synced audio. However, all the case studies and detailed comments on what to watch out for (pun intended) should give you a lot of help in working out what you need for your project.

Good luck,
Joachim

Joachim and Carlos,

Thank you both very much for all your help so far with this problem, all of your input is very constructive. I have had look into the CCTV tech and also that researchgate link, and both have provided me with a few more options. At the moment I am organising a visit to one of the "hackerspaces" near me, to see if anyone can give me some hands-on practical advice, mainly about a suitable power supply solution (as lots of inputs running and recording continuously will reduce the time it can be in-field substantially based upon a standard, small size power supply, as I understand it). I will let you both know when I have managed to develop a prototype, but this may take a month or two as I need to first order a starter kit for Raspberry Pi or Arduino, get my head around the basics, and also contend with my other work. I will be sure to let you all know if I make any headway!

Best wishes, 

Sam

See full post
discussion

Real-time audio monitoring

I am looking for a way to monitor audio in real time instead of writing to an SD card, which means the system gaterhing data needs to stream data back when it is recorded. What...

2 0

Hi,

What you're asking, is it conceptually like turning on the mic on your PC and doing something with the output from the sound card?  Instead of digitising the sound it's possible to do some kind of monitoring in the analogue domain.  It sounds like an interesting problem, tell me more!

Thanks.

What frequecy are you looking to record, how long does the mic need to run for, what is the distance between the microphone and the "listening" site?   As (unless I've missed something) an off the shelf radio microphone does exactly that job.

See full post
discussion

AudioMoth - Big data software?

What are other people using to shift through AudioMoth files to pull out the bats calls - so far I've settled on BatClassify.   If seems to cope with corrupt/zero...

7 0

AH, okay. Yes, if something goes wrong writing to the SD card during a recording, the file won't be closed, and so the file size won't be updated. Let us know if you get any zero length files in †he future since it shouldn't happend. We have another release of the firmware and configuration app coming out in a couple of weeks that makes the routine which writes to the SD card a bit more robust if something goes wrong, uses human readible long file names, and also adds local time to both the configuration app and the resulting time stamps. Alex

Hi David,

Kate Jones suggested trying our their Bat Classifier tool - deep learning tools for bat acoustic signal detection. We've quite a few members here who worked on it if you've got questions (@Rory_Gibb, @snewson, @robin_freeman as a start)

We have a method for finding bat calls that is super quick, free and all code is open source https://t.co/52FPZHLyQU

— Kate Jones (@ProfKateJones) December 21, 2018

Steph

How about using Kaleidoscope and the cluster analysis tool?  Works fine, and is super-quick..

Ta, Carlos

See full post