Bird Song Data Collection App

Design a mobile app to allow users’ to correct machine learning predictions identifying which bird is calling in a recorded audio sample. Bioacoustic expert users will use this app to verify or correct the prediction. 

Objectives

  1. Design a cross-platform mobile app (in Flutter) that retrieves a sound from a cloud database, plays the sound, and records the user response. 
  2. Maintain cloud-based databases of sounds to annotate and of user's annotations previously collected.
  3. Validate product with a small and informal field study. 

Motivations

We seek a crowd-sourced solution to enable domain experts to identify the species of bird present in a short audio clip. The Soundbendor Lab is developing machine learning models to automatically identify the specific species. However, we need human experts to review and correct our predictions, especially those for which our model predicts with a low confidence. 

Qualifications

Minimum Qualifications:
  • interest in bioacoustics

Preferred Qualifications:
  • experience with signal processing
  • CS 492 (mobile) or CS 493 (cloud)


Details

Project Partner:

Patrick Donnelly

NDA/IPA:

No Agreement Required

Number Groups:

1

Project Status:

Accepting Applicants

Website:
https://soundbendor.org/
Keywords:
Mobile DevelopmentClouduser interfaceAudioFlutterBioacoustic
Card Image Capstone