Sirbarksalot Bark Detection in Python
summary
Built app Sirbarksalot detects dachshund barks from a spare webcam, generates spectrograms, and alerts via the Facebook Messenger API. Detection starts with template matching at 0.75 to collect data, yielding 2,000+ six-second clips (Cody 848 barks, Caylee 566), then a simple Keras/TensorFlow CNN refines classification. Runs on a home machine, counts barks per minute, and surfaces false alarms (coffee grinder, gate slams, music). Analysis shows weekday patterns and a nighttime ramp when owners return.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 1:27Project components
- 5:23Motivation and TensorFlow
- 8:11Audio capture and spectrograms
- 10:40Building the model
- 12:48Facebook Messenger alerts
- 13:49Barking patterns analysis
- 16:00False alarms and limits
- 17:44Future improvements
- 18:50Q&A
related talks
21mLeveraging Python for Real-Time Image processing
30mRuby, Red Pandas, and You
40mWho's singing? Automatic bird sound recognition with machine learning
36mbandicoot: a toolbox to analyze mobile phone metadata
23mBirding with Python and Machine Learning
45mThreat Hunting with Python and Pandas
37mHow Python Can Help Detect Early Signals in University Mental Health
33mPython vs Orangutan