It sounds like you are referring to what is often called the "Cocktail Party" problem.  There has been quite a bit of research in this area, but it is a very hard problem.  I would start with the general literature on this topic as there isn't much in the conservation space. Check the hearing aid literature....picking out someone talking in a crowd of voices and sounds.  Humans are pretty amazing at this...machines not so much.  You might consider following up on some recent research at MIT:

https://blogs.nvidia.com/blog/2018/08/28/music-youtube-cocktail-party-problem-ai-artificial-intelligence-deep-learning/

For what its worth, it's a fascinating problem and has a lot of conservation applications.  As Nick mentioned, the biggest challenge is getting enough usable soundscapes.

Hello to you all, I hope everyone is doing ok during this Covid-19 contingency.

Jsut sharing this adaptation for the water proof case for the Audio Moth.

https://www.tetrixecology.com/single-post/Developing-an-Enclosure-for-the-AudioMoth-Acoustic-Logger---Part-1

As well as the points highlighted by Steph, Open Acoustic Devices are also running an online survey to capture the use of AudioMoth. To evaluate the global impact of open-source conservation technology, it is important to know how the technology is being used. This survey will go towards a human-computer interaction study into the usability of tech in conservation, and also used to evaluate the global impact of open-source conservation technology. AudioMoth is being used as a case study to identify best practises for open-source conservation technology development and accessibility. 

It would be great to keep sharing your experiences on this thread and if you have time here with our online survey. 

How do we get notified for the next group buy?

Hi Daniella,

Yes, I have used SWIFT, SongMeter SM4, and Audiomoths. I find Swift to be the most energy efficient. They are also very affordable. Currently, they are my number one choice.  In reality, however, I would say that whether you use Swift over other sensors, particularly Audiomoths, would depend on the type of study you want to do and how often you can get to the sensors to replace batteries. Happy to talk about cons and pros if you want. Just send me an email here: david.fernandez@uwe.ac.uk.

Cheers,

David

Hello Daniella, 

We, WCS Mongolia, used lithium batteries for SWIFT recorders, the most of them still working in until now, since the last August.

At the Cornell SAW, they said alkanium battery will be working 27-44 days, and it depends on configuration sample rate.

We are configured 8K for sampling rate, and 24 hours recording.

Thanks,

Ochiroo

We used the following lithium batteries.

URL:

https://fortebattery.en.made-in-china.com/product/FBDEubqyfGhH/China-C-Size-Lithium-Batteries-Cr26500-Cr26500se-Er26500-Er26500m.html

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 dan@edgeimpulse.com.

Warmly,
Dan