Bird Identification

YouTube ... Quora ...Google search ...Google News ...Bing News

• Capabilities
  • End-to-End Speech ... Synthesize Speech ... Speech Recognition ... Music
  • Video/Image ... Vision ... Colorize ... Image/Video Transfer Learning

Researchers at the Cornell Lab of Ornithology are passionate about studying birds and biodiversity and advancing conservation. Their mission encompasses fieldwork, laboratory research, data-intensive science, student training, and globally renowned citizen-science and lifelong learning programs

Merlin

• Merlin
  • Merlin | Google Play
  • Merlin | Apple Store

Merlin Bird ID is a mobile app developed by the Cornell Lab of Ornithology that uses machine learning and computer vision to identify bird species from photos and sounds. Merlin's approach to sound identification uses AI technology powered by tens of thousands of citizen scientists who contributed their bird observations and sound recordings to the Lab's Macaulay Library via eBird, the Cornell Lab's global database¹. The app is given a training set of bird vocalizations and photos whose identity is known. The program then seeks to teach itself how to identify each of those vocalizations or images. Once trained, the program is then given unknown vocalizations or images and provides an identification. The programmer can check whether those identifications are correct. If not, the program is informed of the wrong decision and will revise its learning. So, the application should get better and better over time.

Now, using the vast photo library at the Cornell Lab of Ornithology, the app is well trained. Over 8,000 species can be identified now, 80% of the world’s bird species. Some users report a 90% or higher rate of correct identifications. Merlin gets over 90% of the sound identifications right. The app is available for free on Android and iOS platforms.

Source: Conversation with Bing, 6/11/2023 (1) Merlin and the Use of Artificial Intelligence in Bird Identification .... https://web.colby.edu/mainebirds/2022/09/11/merlin-and-the-use-of-artificial-intelligence-in-bird-identification/. (2) Merlin Bird Photo ID - Cornell Tech. https://www.tech.cornell.edu/built/merlin-bird-photo-id/. (3) What Bird is Singing? Ask the Merlin Bird ID App for an ... - Newswise. https://www.newswise.com/articles/what-bird-is-singing-ask-the-merlin-bird-id-app-for-an-instant-answer. (4) What bird is singing? Merlin Bird ID app offers instant answers. https://cals.cornell.edu/news/2021/06/what-bird-singing-merlin-bird-id-app-offers-instant-answers.

Photo ID: Merlin Photo ID uses computer vision technology, developed as part of Dr. Grant Van Horn’s doctoral work at Caltech, to identify birds in photos. Photo ID was developed in collaboration with Dr. Pietro Perona’s computational vision lab at Caltech, and Dr. Serge Belongie’s computer vision group at Cornell Tech, collaborators on the Visipedia project. First publicly released Nov 30th, 2017.

Sound ID Sound ID uses recordings archived in the Macaulay Library to learn how to recognize the vocalizations of different bird species. Sound ID is trained on audio recordings that are first converted to visual representations (spectrograms), then analyzed using computer vision tools similar to those that power Photo ID. Dataset preparation began in 2020 with model development starting in early 2021. Sound ID was developed in-house at the Cornell Lab of Ornithology, led by Dr. Grant Van Horn with assistance from Dr. Benjamin Hoffman. S We thank the many annotators that helped curate hundreds of audio recordings for each species. First publicly release June 23rd, 2021.

eBird

eBird | Cornell Lab of Ornithology

eBird began with a simple idea—that every birdwatcher has unique knowledge and experience. Our goal is to gather this information in the form of checklists of birds, archive it, and freely share it to power new data-driven approaches to science, conservation and education. At the same time, we develop tools that make birding more rewarding. From being able to manage lists, photos and audio recordings, to seeing real-time maps of species distribution, to alerts that let you know when species have been seen, we strive to provide the most current and useful information to the birding community.

eBird is among the world’s largest biodiversity-related science projects, with more than 100 million bird sightings contributed annually by eBirders around the world and an average participation growth rate of approximately 20% year over year. A collaborative enterprise with hundreds of partner organizations, thousands of regional experts, and hundreds of thousands of users, eBird is managed by the Cornell Lab of Ornithology.