Bird sounds identification - a new approach? (2 Viewers)

[Peter.home]

I have an idea for the next time I'm visiting a region where I'm totally unfamiliar with the bird calls.

It would be to download or obtain calls for all the regular species, as individual files, then add the starting call frequency to the beginning of the file.

E.g. 1500 speciesX.wav

I'd then order them in ascending frequency, and on hearing a call, I could fast forward thru the playlist until I got the starting frequency correct, then work out the call from the small number of recordings in the approximate frequency.

I suspect it would be a pain to set up, maybe 10 a day, you'd have a library of 100 species in a couple of weeks.

To me, it would be the obvious, easiest way of identifying bird calls, given merlin seems quite slow to be rolling out across the world!

Any thoughts?

 

[davpen]

Hmm, maybe, but you only have to think about Meadow Pipits and the wide range of frequencies their calls can start at.

 

[Peter.home]

The other issue, would I end up with Bittern, Nightjar, then pretty much everything else squashed into a very narrow frequency range? (Using UK as an example)

The trouble is every other reference or method requires you to go to a species, then listen to the call. Was it that? If not, then you aren't getting any closer to the answer.

There is nothing that takes the call and leads you towards the correct species.
Apart from Merlin, which would be the game changer if it was more widely rolled out.

 

[Jon.Bryant]

I was recently in Ecuador and quite stunned about how many species had really high frequency tinny voices. I think if you arranged calls by frequency you would end up with a load of recordings grouped at the top end of the frequency range.

There are other ways to categorise sounds. The Peterson Sound Guide (for North America) describes a lot different sounds - up slurs, down slurs, trills etc. so perhaps you could use this to refine your list - I.e 4KHz containing up slur elements. But this would be an awful lot of work.

Probably best just to sort recordings by how common the species are and then trying to learn the sounds…. Or employ a good local guide who knows the calls of the target species - the guides we had a n Ecuador were all very knowledgeable on bird callls.

 

[LittleBitOfBreadNoCheese]

Harmonics are the other big killer to the plan I fear. Meanwhile birdNET has 6k species it can handle so on foreign trips may be more useful, although you either need internet in the field or save and analyse later at the hotel etc.