Observed arrival · 2026-10-10
SoraLeuweung: a forest-listening LoRa alert demo
An interactive presentation of an offline acoustic-AI system intended to detect illegal logging and alert forest rangers in areas without cellular service.
- For
- Forest rangers and conservation-technology teams
- Worth noticing
- Its stated design pairs 16 kHz audio sensing with an ESP32 and encrypted LoRa alerts, with a listed SRAM footprint of 2.6 KB.
Field notes
The proposed sensing node combines an INMP441 microphone with an ESP32, extracting MFCC features and running quantized inference locally. The page specifies a 16 kHz audio stream, 2.6 KB SRAM use, and AES-128-encrypted LoRa messaging; its browser interface also displays staged telemetry and a sample classification probability. These are presented as design details and simulation material, not as independently verified field results.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue