- 3DRAG + MOTORFISH: How to prevent step losses in 3D printerPosted 2 weeks ago
- Motorfish: The stepper does not misses a step (the Firmware)Posted 3 weeks ago
- Motorfish: The stepper does not misses a stepPosted 4 weeks ago
- “Maker Faire Rome – The European edition” returns to the Gazometro OstiensePosted 3 months ago
- Single-chip saturation meterPosted 3 months ago
- Raspberry Pi Relay BoardPosted 5 months ago
- PCB CAD, A SELECTION GUIDEPosted 5 months ago
- Automatic dispenser for catsPosted 6 months ago
- Do you have a mask? RASPBERRY Pi SEES ITPosted 6 months ago
- RFID based Attendance system using Arduino and External EEPROMPosted 6 months ago
Bird Sound Classifier
The project attempts to recognize different bird calls by continuously listening to audio through the Nano 33 BLE Sense’s built-in microphone. The call of the bird heard will be analyzed and classified; if it is not heard, the audio will be classified as background noise. This project can be useful for people interested in birding or those who want to understand the patterns of calls.
Edge Impulse fully supports Arduino Nano 33 BLE Sense, a compact development board containing a Cortex-M4 microprocessor, motion sensors, a microphone and BLE.
After configuring the Nano 33 BLE sensor with the Edge Impulse framework, we can continue with the next step, which is to build a machine learning model.
Because we need a lot of bird data and it’s hard to find high quality and quantity, we got it from Xeno-Canto, which is a large database dedicated to sharing bird sounds around the world.
We downloaded about 20-25 audio files for each bird and worked on the preprocessing using a software called Audacity.
After configuring the neural network model and doing some tests, we got the following results: