Eurasian coot heading south? 89%
Back to Listening
Bladerdak listens to
§ B · Bladerdak Acoustics · north/south
Bladerdak listens to
what happens up high.
The microphones sit in the attic, pointing out from under the gutters towards the north and the south. From here we hear the treetops, the cranes passing over, the woodpeckers tapping out their territory. Since January we also know from which direction each bird is singing.
About the confidence score
An algorithm is no ornithologist
Every detection comes with a confidence score. 92% chance it's a blackbird, 23% chance it's a noctule. The number looks scientific, but it remains a model's guess. Often very good. Sometimes completely off.
A few from this garden. A bicycle bell once became a bullfinch. A bouncing ball down the street, an eider. One pipistrelle sounds almost identical to another, as far as BatDetect2 is concerned. The algorithm misfires there regularly, even at 95%. Rare species are extra suspect. The model has seen fewer of them, so a detection like that stands on shaky ground.
I check the less common and rare species myself. For each detection I play the recording back and see if it holds up. But that leaves me running behind sometimes. I might only get to it days later, and until then a doubtful case just sits on the site.
So those numbers are a hint, not an identification. An algorithm that listens day and night hears more than I ever could on my own, and that's already something.
Latest detection
★ Hoogtepunthelderste opname die hier is geregistreerd
§ B-02 · Live on the roof From the microphones in the attic, aimed at the treetops
Just in.
Every detection from BirdNET-Pi Bladerdak appears here right away. A pause in the stream means no activity, not that the pipeline has stopped.
connecting
-
Listening.
No detections since you opened this page. A quiet stretch means nothing was heard, not that the pipeline is down.
§ B-04 · What are we actually hearing Vocalisation types, top 10
Birds do not just sing. They call, they alarm and they keep in contact.
Our own CNN models decide, per detection, whether it is song, call, alarm or something else. Tally of the last seven days, for the top 10 most-heard species.
| Species | song | call | alarm | other |
|---|---|---|---|---|
| Common swift | 8,603 | 5,440 | 8 | 2,687 |
| Common blackbird | 3,358 | 1,572 | 662 | 2,262 |
| Common wood pigeon | 492 | 1,062 | 124 | 297 |
| Western jackdaw | 560 | 491 | 76 | 264 |
| Eurasian blue tit | 62 | 944 | 68 | 277 |
| Great tit | 781 | 238 | 47 | 174 |
| Eurasian magpie | 269 | 113 | 342 | 98 |
| Common chaffinch | 285 | 139 | 149 | 150 |
| European robin | 162 | 58 | 121 | 93 |
| Eurasian collared dove | 8 | 323 | 4 | 44 |
Song = territorial (song/subsong/display song). Call = communication (call/begging/flight call/nocturnal migration call). Alarm = predator spotted. Other = raw drumming, or not classified by the model.
§ B-05 · Listening Top 3 per species, top 9 on confidence
Highlights, automatically selected.
Every night a script makes a fresh selection: the three clearest recordings per species from the past year. No quality curation, only statistics. Confidence above 85 per cent.
9 soorten · 27 fragmenten getoond · nog 95 soorten in het archief. Selectie gemaakt op 1 jun 26.
Alle soorten bekijken → § B-06 · Highlights · all time Hand-picked, with personal context
The classics.
Rarities, one-off passers-by, and moments I do not want to forget. Each one fished out of the archive by hand.
-
testkeep
Bramblingtest
-
Common blackbird
Blaffende hond en Merelzang
§ B-07 · Today versus normal What stands out?
- Today 2 different species, with a total of 27 detections.
- Peak today around 00:00 (27 detections). The average over the last thirty days peaks around 08:00.
- Today fewer detections than the average over the last thirty days (-99%).
Detections per hour 0–23 h
today
30-day average
§ B-08 · Under the hood How Bladerdak is built
Bladerdak runs on a Raspberry Pi 5 in the attic. Two microphones hang under the gutters, plugged into a Steinberg UR22mkII audio interface. One points north, one points south. BirdNET-Pi analyses the audio every three seconds. A second layer (197 of our own CNN models) decides whether it is song, call or alarm. From the difference in loudness between the two channels we know whether the bird is north or south of the house.