BirdNET

Every birder knows the feeling. A song comes out of the canopy, it’s clearly a bird, and you can’t see it. For most of birding’s history, the answer was years of practice, a stack of audio guides or a more experienced friend. BirdNET aims to replace all three with the microphone already in your pocket.

BirdNET comes from the BirdNET research project, run jointly by the K. Lisa Yang Center for Conservation Bioacoustics at the Cornell Lab of Ornithology and Chemnitz University of Technology in Germany. That matters. This isn’t a startup chasing subscriptions or a hobby project that will be abandoned in eighteen months. It’s an academic tool built on neural networks trained on large collections of bird recordings, and it can recognise more than 3,000 of the world’s most common species by their songs and calls.

Its market position is unusual. BirdNET’s most direct rival is Merlin, which comes from the same Cornell Lab. Merlin is the consumer product: polished, guided and built for real-time listening. BirdNET is closer to the research instrument with a mobile front end. You record, you select the relevant part of a spectrogram, and you submit it for analysis. The app returns its best guess with a confidence level. Your location and the date narrow that guess to birds that should plausibly be around you.

The business model is easy to explain because there isn’t one. No ads, no account, no paywall. The trade is a different kind of value: each submission adds to a dataset that researchers use to study bird distribution, migration and conservation. You get an identification, and science gets a data point.

The store rating sits at 4.0 out of 5 from 215 ratings. That points to a tool people trust, not one they love. That gap is what this review is about.

• AI-Powered Sound Identification: Neural networks trained on large bird-audio datasets identify more than 3,000 common species worldwide from songs and calls.
• Spectrogram Selection: You don’t just hold up the phone and hope. You visually isolate the call on a spectrogram before submitting it, which gives you precise control over what gets analysed.
• Confidence Scoring: Each result comes with a confidence level, so the app tells you how sure it is instead of presenting every guess as fact.
• Location and Date Filtering: It uses your GPS position and the calendar date to weight results toward species expected in your area at that time of year.
• Personal Observation Log: Saved identifications include species and location data, so over time you build a personal life list of birds you’ve heard.
• Citizen Science Contribution: Every submission feeds research on bird distribution, migration and conservation.
• Zero Monetisation Friction: No ads, no sign-up and no in-app purchases.

The Core Loop: Record, Select, Submit

BirdNET’s workflow is deliberate, and that’s both its main strength and its main weakness. You hear a bird, open the app and record. The app then shows you a spectrogram, a visual map of sound with frequency on one axis and time on the other. You drag a selection box around the part that holds the call and submit it.

For a beginner, this is real onboarding friction. Most people have never read a spectrogram, and the first few attempts must feel like being handed an oscilloscope and told to have fun. Which streak of grey is the robin? Which is the traffic? The app doesn’t do much to teach you.

Stick with it, though, and something interesting happens. The spectrogram becomes a teacher. Over time a user starts recognising shapes: the rising whistle of one species, the dense buzzy block of another. Without meaning to, they’ve learned a skill ornithologists use professionally. Merlin’s real-time listening is more convenient, but it puts the user in the passenger seat. BirdNET makes you drive, and drivers learn the roads.

That’s the app’s central idea, and it reads as partly intentional. You are an active participant in the identification, not just someone tapping a button for a magic result.

Accuracy: Strong Where It Counts, Brittle at the Edges

On common species with a clean recording, BirdNET should be reliable — that’s what you’d expect from a model trained on a large dataset dominated by frequently recorded birds. The dawn chorus in a suburban park or a single songbird on a fence post is BirdNET’s home ground.

The trouble starts when conditions get messy. Noisy recordings degrade accuracy noticeably. Wind, traffic, overlapping birds and distant calls all push the model toward lower confidence or outright mistakes. Rare and uncommon species are the other weak point. This follows from how machine learning works: a model can only be as good as its training data, and uncommon birds are, by definition, under-recorded.

The confidence score is the safeguard here, and it’s the most important number on the screen. A high-confidence result on a common local species deserves trust. A low-confidence result on something unexpected should be treated as a hypothesis, not an identification. Experienced users read it that way. Beginners may not, and the app could do much more to explain what a 40% match actually means.

Location and date filtering helps a lot. Weighting results toward species that should be present in your region and season cuts down on absurd suggestions, such as a tropical species in a Minnesota winter. It’s a simple contextual prior, and it does a lot of work behind the scenes.

The Citizen Science Bargain

The reason to choose BirdNET over a more polished option is the research it supports. Every confirmed observation becomes a point in a global dataset on how birds move, where populations are shifting and how species respond to habitat and climate pressure. It turns a solitary hobby into a contribution, and unlike most “your data helps us improve” pitches, this one has a clear and worthwhile beneficiary.

It also explains the no-account, no-ads design. The project doesn’t need to monetise you. It needs your recordings. That’s an unusually honest arrangement in the app economy.

The Competitive Problem

BirdNET’s real challenge comes from its own family. Merlin, from the same Cornell Lab, offers a more consumer-friendly experience with more polish. BirdNET Live, the project’s newer on-device app, now covers what some users relied on the original for. That leaves the classic BirdNET app in an awkward spot. It’s still a good tool, but its niche is shrinking. Its strongest remaining audience is people who want the deliberate spectrogram workflow and the closer link to the research project.

BirdNET looks like what it is: a research tool with a mobile shell. The interface is functional but dated. The layout is utilitarian, the type is plain and there’s little visual hierarchy to guide a newcomer. It has none of the refined illustration, smooth transitions or careful information design that set the standard for nature apps today.

To be fair, the spectrogram view is clear and readable, and that’s the screen that matters. Selection is precise enough for the job. The observation list is basic but does its work. Nothing here is broken. It just doesn’t look like anyone has spent real money on product design, because no one has.

Performance is mostly a matter of turnaround. The classic workflow is submission-based, so results depend on that step instead of appearing in real time as you listen. That suits deliberate identification and is a poor fit for the “what’s that, quick, before it stops” moment. Anyone who wants instant on-device feedback is the audience BirdNET Live and Merlin were built for.

On iOS and Android, the core experience is functionally the same, and both platforms carry the same 4.0 store rating. The free, no-cost model fits budget-conscious students and markets where Android dominates.

Touch suits the spectrogram selection well. Dragging a box around a call with a finger is intuitive once you know what to look for, though finer selections on smaller phones can feel imprecise. A larger-screen phone helps a lot, both for reading the spectrogram and for making clean selections. Both platforms depend on the hardware microphone, so recording quality varies between devices. A phone with good noise handling will give the model cleaner input, and that directly improves accuracy.

The app doesn’t do much with platform-specific features. Don’t expect deep widget support, polished system integrations or anything that feels tuned to either ecosystem. It’s a cross-platform research client, and it acts like one.

BirdNET is the most honest app in its category. It doesn’t flatter you or hide its uncertainty, and it doesn’t try to sell you anything. It gives you a capable neural network, a spectrogram and a confidence score, then trusts you to use them well. For common birds in reasonable conditions, it delivers.

That honesty has a cost. The interface is dated, the learning curve is real, and accuracy falls off exactly where a beginner most needs help: noisy environments and unfamiliar species. Merlin is the friendlier choice for casual users, and BirdNET Live has taken over part of this app’s job.

Still, the spectrogram is the reason to come back. Most AI tools make you more dependent on them. BirdNET, almost by accident, makes you less dependent over time, because it teaches you to see sound. Combine that with a zero-cost, zero-ads model and a real contribution to conservation science, and we recommend it to the right audience — birders who want to learn by ear — even if it isn’t the first choice for a fast ID.