How can computers learn to recognise birds from sounds? The Cornell Lab of Ornithology and the Chemnitz University of Technology are trying to find an answer to this question.
Their research has led to the development and evolution of BirdNET – an artificial neural network that has learned to detect and classify avian sounds through machine learning.
This webinar provides an introduction to BirdNET, how it works and and how the use of BirdNET can be scaled to generate huge biodiversity datasets (with a case study from Wilder Sensing).
BirdNET: Advancing Conservation with AI-Powered Sound ID
Dr Stefan Kahl (Cornell Lab of Ornithology and Chemnitz University)
Learn about how BirdNET was built with Dr Stefan Kahl. He’ll cover some basics about AI for sound ID, present a few case studies that have used BirdNET at scale and then conclude with some thoughts on the future of AI in bioacoustics.
Dr Stefan Kahl is a computer science post-doc at the Cornell Lab of Ornithology and Chemnitz University in Germany, working on AI algorithms for animal sound ID. He is lead of the BirdNET team, where they develop state-of-the-art tools for anyone interested in large-scale acoustic monitoring.
How did you handle species with multiple distinct call types?
We put all of the files from one species into a folder and told the AI model that this is one species. That has worked relatively well, these models are able to distinguish different call types for the same species so we can do calls, we can do songs, we can do all kinds of call types. We know that these feature vectors embed these call types, and we can re-construct it. So, after you’ve run BirdNET, instead of the class scores if you take these embeddings, you can do a clustering and cluster out these different call types and you’ll end up with a nice visualisation.
Sometimes what people will do, if they are interested in something specific like mating calls, you can train a custom classifier, as long as it is a category. So, if you can categorise it, it can become a model. So, if you want to run a call type model instead of a species model you can. One category of call types which is challenging is flight calls. Not all of the species have sufficient flight call data and flight calls tend to be short. I would exclude those.
How did you settle on the three second segments for the sampling?
We wanted to reduce the input size as these models are computationally expensive. So, the bigger the input, the more computationally expensive, so we want the smallest spectrogram you can have that still retains all the detail.
During my PhD, I did an empirical study looking at the average bird song length and the literature gave a time of two seconds. I added half a second before and after to have some wiggle room and that’s why we chose three seconds. We know some people are using five seconds and that is usually fine if you have longer context windows that might help with call sequences. Sometimes three seconds is not good enough as you need a temporal component to it.
Do recordings need to be added to Xeno-canto or can you access recordings from Merlin and other systems?
Merlin doesn’t currently leverage user’s observations, i.e., Merlin is not collecting data that we can use. We have access to recording submitted to eBird and Macaulay Library, though Xeno-Canto is a bit faster to allow non-bird uploads. The best The best way for users of BirdNET-Pi is BirdWeather (get a device ID and hook it up to the BirdWeather platform).
Is there a way to reduce the number of false positives in BirdNET?
Yes, you can also use false positives to train a custom classifier and then BirdNET will (hopefully) learn to separate target from non-target sounds. So basically, using those “negative” samples to train a model.
What is needed to scale BirdNET fast?
More data. Plus anecdotal evidence on how people are using it, so we can learn what we need to tackle to make it more useful.
Do you know if anyone is using BirdNET to look at the relationship between anthropogenic noise and species abundance?
Not abundance but disturbance – there’s a project going on in Yellowstone National Park looking at the impact of snowmobiles on bird vocalisation. They found that engine sounds are a significant disturbance and needs to be managed, i.e., birds will stop vocalizing for extended periods of time, even well after the snowmobiles have passed.
What data augmentation techniques do you use (if any) to expand your training dataset?
MixUp (mixing multi recordings into one) is the most important and most effective augmentation. We had a student look into augmentations a while ago.
A Scalable Platform for Ecological Monitoring
Lorenzo Trojan (Wilder Sensing)
How can we measure the impact of wildlife restoration, assess biodiversity loss, and evaluate the effectiveness of environmental policies? Passive bioacoustic monitoring offers a powerful solution, enabling continuous, large-scale coverage of ecosystems. To fully harness its potential, we need a scalable, robust software platform capable of handling vast audio datasets, detecting key biological sounds using AI, and extracting ecological insights such as species richness and assemblage trends.
This presentation explores the challenges of biodiversity monitoring with passive audio recorders, the processing and analysis of large-scale acoustic data, and the technologies that make this approach viable and impactful. We’ll also showcase how the Wilder Sensing Platform is purpose-built to meet these demands—providing researchers, conservationists, and policymakers with an intuitive, scalable, and efficient tool for biodiversity monitoring and ecological surveys.
Dr Lorenzo Trojan is a technologist with a PhD in Astrophysics and over a decade of leadership experience in high-growth tech startups. His expertise spans remote sensing, cloud computing, DevOps, and AI. As CTO of Wilder Sensing, he leads the development of a scalable platform for ecological monitoring, driven by a commitment to innovation, inclusivity, and impact.
What is the trade-off between sample rate and bits per sample and the accuracy of detection as a method for reducing storage requirements?
From a series of preliminary experiments we conducted using recordings encoded at different sampling rates, we did not find a substantial drop in the quality or quantity of detections obtained from BirdNET.
These results are preliminary and based on a limited dataset; however, they provide a clear indication that reducing sampling rate can be an acceptable approach to mitigating data volume and storage costs with little or negligible impact on detection quality.
The Wilder Sensing monitoring guidelines are therefore based on these findings, adopting a configuration that represents a practical compromise between detection performance and data volume.
What are the main features within the Wilder Sensing platform that make BirdNET more accessible for people with boots on the ground?
Our platform enables anyone requiring biodiversity monitoring of large estates to streamline the process of collecting, storing, retrieving and analysing the large volumes of data associated with passive bioacoustics, providing a simple and accessible way to understand what species are present within a landscape.
Audio files are collected and organised into projects and linked to individual recording devices, with metadata such as recording time and geolocation automatically associated with each recording. The recordings are automatically processed using BirdNET, and the resulting detections are presented through simple and accessible analytics dashboards that allow users to explore species activity across locations and time periods. Audio files can also be optionally stored in long-term archives for auditing and verification purposes.
How affordable is large-scale bioacoustics monitoring for organisations with limited funding?
Passive acoustic monitoring is one of the most cost-effective approaches to biodiversity monitoring. It enables large estates and diverse ecosystems to be monitored continuously over extended periods using relatively inexpensive recording devices and minimal field manpower. This makes it particularly well suited to organisations with limited monitoring budgets, such as NGOs, academic institutions and conservation groups.
Wilder Sensing further reduces the operational complexity associated with large-scale bioacoustic monitoring by providing a platform for storing, organising and analysing the large volumes of audio data generated by these deployments. The platform allows users to perform automated BirdNET analysis, explore results through intuitive analytics tools, and share datasets and findings without needing to manage specialised infrastructure or data pipelines.
As part of our mission to support scientific understanding of the environment, Wilder Sensing offers pricing structures designed to remain accessible to research and conservation organisations.
What are the costs outside of Wilder Sensing that not-for-profits need to consider in their budget?
Outside of the Wilder Sensing platform itself, the main costs associated with a bioacoustic monitoring project typically relate to field equipment and operational logistics. These include the purchase of recording devices, memory cards, and accessories required for deployment such as protective enclosures, mounting hardware or fencing where necessary.
Field operations also represent an important component of the overall budget. Organisations need to consider the staff time required for deploying devices, visiting monitoring sites to replace batteries and SD cards, and retrieving equipment at the end of the monitoring period. Additional ongoing costs may include replacement batteries, maintenance of devices, and the occasional replacement of damaged or lost equipment.
The relative importance of these costs can vary considerably depending on the nature of the monitoring project. Factors such as project duration, site accessibility, ecological objectives and the availability of trained staff or volunteers can significantly influence the total operational cost.
A long-term monitoring project lasting several months or years may require regular site visits to replace batteries and memory cards. If monitoring locations are difficult to access or require specific safety procedures or permissions (wetlands, railway corridors, renewable energy installations) additional logistical planning may be required.
In contrast, shorter projects targeting specific species during limited periods (for example, during breeding or nesting seasons) may require fewer devices, fewer site visits and lower operational costs overall.
In some cases, Wilder Sensing may also be able to support projects through bespoke monitoring arrangements. While this is not part of our standard service offering, we have occasionally worked with organisations—particularly academic institutions—to provide tailored monitoring packages where recording equipment is temporarily loaned as part of the project. These arrangements are considered on a case-by-case basis and are intended to help facilitate research or conservation initiatives where budget constraints might otherwise limit the feasibility of large-scale monitoring.
Useful links
- BirdNET website: https://birdnet.cornell.edu/
- BirdNET GitHub repository: https://github.com/birdnet-team
- BirdNET Analyzer documentation: http://birdnet.cornell.edu/analyzer
- Identifying Bird Sounds with the BirdNET Mobile App: https://www.youtube.com/embed/f144CSEoYuk
- What’s that bird song? ID birds by sound with BirdNET: https://www.youtube.com/embed/MQHunTLt1TI
- Kahl et al. (2021) BirdNET: A deep learning solution for avian diversity monitoring: https://www.sciencedirect.com/science/article/pii/S1574954121000273
- Pérez-Granados (2023) BirdNET: applications, performance, pitfalls and future opportunities: https://onlinelibrary.wiley.com/doi/full/10.1111/ibi.13193
- Wilder Sensing website: https://wildersensing.com/
- Sign up for the Wilder Sensing e-newsletter: https://2e428x.share-eu1.hsforms.com/2XxP8d_6lRSmBIKH7uwruXQ
- Contact Wilder Sensing: https://wildersensing.com/contact/
Wilder Sensing ecoTECH blogs
- How Can We Use Sound to Measure Biodiversity: https://biologicalrecording.co.uk/2024/07/09/bioacoustics-1/
- Can Passive Acoustic Monitoring of Birds Replace Site Surveys blog: https://biologicalrecording.co.uk/2024/09/17/bioacoustics-2/
- The Wilder Sensing Guide to Mastering Bioacoustic Bird Surveys: https://biologicalrecording.co.uk/2024/11/26/bioacoustics-3/
- Bioacoustics for Regenerative Agriculture: https://biologicalrecording.co.uk/2025/03/31/bioacoustics-for-regen-ag/
- AI-powered Bioacoustics with BirdNET: https://biologicalrecording.co.uk/2025/07/08/birdnet/
- Making the Most of Bird Sounds: https://biologicalrecording.co.uk/2026/03/11/making-the-most-of-bird-sounds/
Event partners
This blog was produced by the Biological Recording Company in partnership with Wilder Sensing, Wildlife Acoustics and NHBS.
- Sign up for the Wilder Sensing e-newsletter: https://2e428x.share-eu1.hsforms.com/2XxP8d_6lRSmBIKH7uwruXQ
- Wildlife Acoustics Song Meter Micro 2: www.nhbs.com/song-meter-micro-2
- Wildlife Acoustics Song Meter SM5: https://www.nhbs.com/song-meter-sm5-acoustic-recorder
- Check out the NHBS Field Guide Sale: www.nhbs.com/spring-promotions
More for environmental professionals
Biological Recording 
Impressive work! BirdNET shows how AI can truly help nature. Excited to see where it goes next.
LikeLike