M
MoltPulse
regg (ronibandini/reggaetonBeGone) | MoltPulse
regg Detects reggaeton genre with Machine Learning and sends packets to disable BT speakers (hopefully)
Based on repository activity, growth velocity and community engagement.
28
Is this actually an AI agent?
π§π« Reggaeton Be Gone
Edge Machine Learning device that recognizes reggaeton and triggers a Bluetooth test routine on a configured speaker
Roni Bandini β Buenos Aires, Argentina β February 2024
Reggaeton Be Gone is an experimental Raspberry Pi device inspired by TV-B-Gone .
It continuously samples ambient audio, uses an Edge Impulse audio-classification model to determine whether reggaeton is playing, displays the inference confidence on a 128Γ32 OLED, andβwhen the configured threshold is exceededβtriggers a Bluetooth test routine against a configured speaker.
The original motivation was simple: a neighboring Bluetooth speaker playing loud reggaeton every morning.
The resulting project combines:
ποΈ Audio acquisition
π§ Edge Machine Learning
π΅ Music-genre classification
π Raspberry Pi
π‘ Bluetooth experimentation
π₯οΈ OLED feedback
π Event logging
π§· Simpler Alternative: Pocket Gone A later project explores the Bluetooth side without Machine Learning or a Raspberry Pi.
Pocket Gone is smaller, cheaper and portable:
The evolution from Reggaeton Be Gone to Pocket Gone was also presented at Nerdearla 2025:
ποΈ Architecture flowchart LR
MUSIC["π΅ Ambient Music"]
MIC["ποΈ USB Microphone"]
PI["π Raspberry Pi 3"]
EI["π§ Edge Impulse<br/>AudioImpulseRunner"]
CLASS{"reggaeton?"}
OLED["π₯οΈ 128Γ32 OLED"]
LOG["π log.txt"]
BT["π‘ Bluetooth Test Routine"]
SPEAKER["π Configured Speaker"]
MUSIC --> MIC
MIC --> PI
PI --> EI
EI --> CLASS
CLASS --> OLED
CLASS --> LOG
CLASS -->|"score > threshold"| BT
BT --> SPEAKER
The ML inference runs locally on the Raspberry Pi using an Edge Impulse .eim deployment.
β¨ Features
π΅ Reggaeton / other-music classification
π§ Edge Impulse Linux inference
ποΈ USB microphone or USB audio interface
π₯οΈ 128Γ32 SSD1306 OLED
π Physical start button
π― Configurable inference threshold
π‘ Configurable Bluetooth target
π Timestamped activity log
π¬ Multiple experimental Bluetooth methods
π§ͺ Manual forceFire test mode
πΌοΈ Custom OLED graphics and font
𧱠3D-printed front panel
π Several later experimental versions
π§ Machine Learning The original model was trained because common music-genre datasets such as GTZAN did not provide a dedicated reggaeton class.
flowchart LR
SONGS["π΅ Music Samples"]
WAV["WAV<br/>Mono / 16 kHz"]
SPLIT["βοΈ 4 s Windows"]
MFE["π¬ MFE"]
NN["π§ Classification"]
EIM["π¦ Linux ARM .eim"]
PI["π Raspberry Pi"]
SONGS --> WAV
WAV --> SPLIT
SPLIT --> MFE
MFE --> NN
NN --> EIM
EIM --> PI
Original settings documented for Version 1:
| Parameter | Value |
| ----------------- | -------------: |
| Audio | Mono WAV |
| Sample rate | 16 kHz |
| Window size | 4000 ms |
| Processing | MFE |
| Learning block | Classification |
| Target deployment | Linux ARM |
| Runtime format | .eim |
The current source expects:
model = "reggaetonbgone-linux-armv7-v4.eim"
π€ Public Model A public reference model is now available on Hugging Face:
The model card specifies:
Audio: 16 kHz WAV / mono
Labels:
- reggaeton
- otros
The published reference model was trained with only six songs , so the model card recommends training a larger dataset for improved generalization.
This is particularly important when deploying the classifier in rooms with:
speech
television
other music genres
different speakers
background noise
varying microphone distance
ποΈ Edge Impulse Linux Runner The Python application uses:
from edge_impulse_linux.audio import AudioImpulseRunner
and initializes the model with:
with AudioImpulseRunner(modelfile) as runner:
model_info = runner.init()
labels = model_info['model_parameters']['labels']
Audio is continuously classified through:
for res, audio in runner.classifier(
device_id=selectedDeviceId
):
The configured USB input is:
Change this value to match the desired microphone/audio interface.
π― Detection Threshold Version 1.0 currently uses:
The program specifically watches the:
Below the threshold, the OLED displays:
if label == 'reggaeton' and score <= threshold:
updateScreen(
"Is reggaeton?",
str(round(score * 100, 2)) + " %"
)
When the threshold is exceeded:
if label == 'reggaeton' and (
score > threshold or forceFire == 1
):
the configured Bluetooth test action is triggered.
βοΈ Version 1.0 Configuration The principal settings are grouped near the beginning of:
myPath = "/home/pi/reggaeton/"
selectedDeviceId = 1
method = 1
targetAddr = ":::::"
packagesSize = 800
threadsCount = 1000
threshold = 0.95
myDelay = 0.1
forceFire = 0
model = "reggaetonbgone-linux-armv7-v4.eim"
| Setting | Purpose |
| ------------------ | --------------------------- |
| myPath | Runtime asset directory |
| selectedDeviceId | Audio input |
| method | Bluetooth experiment mode |
| targetAddr | Test speaker address |
| packagesSize | Method-specific parameter |
| threadsCount | Number of repetitions |
| threshold | ML confidence threshold |
| myDelay | Delay between actions |
| forceFire | Skip ML trigger for testing |
| model | Edge Impulse .eim |
The source code is intended for controlled experimentation with hardware you own.
π Physical Start Button The application does not immediately begin audio classification.
GPIO.setmode(GPIO.BCM)
buttonPin = 26
GPIO.setup(
buttonPin,
GPIO.IN,
pull_up_down=GPIO.PUD_UP
)
until GPIO 26 is pulled LOW.
After the button is pressed:
and audio inference begins.
π₯οΈ OLED Interface The build uses a monochrome SSD1306 128Γ32 OLED .
disp = Adafruit_SSD1306.SSD1306_128_32(
rst=None
)
The screen displays information such as:
Reggaeton BeGone
Listening...
Reggaeton BeGone
Is reggaeton?
92.61 %
The UI is rendered using Pillow:
from PIL import Image
from PIL import ImageDraw
from PIL import ImageFont
font = ImageFont.truetype(
'whitrabt.ttf',
12
)
π Wiring
OLED | OLED | Raspberry Pi |
| ---- | ------------ |
| SDA | GPIO 2 |
| SCL | GPIO 3 |
| VCC | Power |
| GND | GND |
and verify the display with:
Push Button | Button | Raspberry Pi |
| ------ | ------------ |
| Pin 1 | GPIO 26 |
| Pin 2 | GND |
The internal pull-up is enabled in software.
Audio USB audio interface + microphone
The original build used a Behringer Xenyx 302USB .
π οΈ Hardware | Component | Quantity |
| ------------------------------------------------------------------------------ | -------: |
| Raspberry Pi 3 | 1 |
| DFRobot 128Γ32 OLED | 1 |
| DFRobot Push Button | 1 |
| USB microphone / USB audio interface | 1 |
| microSD card | 1 |
| 5 V / 3 A power supply | 1 |
| Female-female jumper wires | Several |
| Bluetooth speaker for controlled testing | 1 |
| Custom enclosure/front panel | 1 |
An optional external Bluetooth adapter can also be used for experimental versions.
π Runtime Flow flowchart TD
START["Power On"]
UI["Display Target + Method"]
WAIT["π Wait for Button"]
LOAD["π§ Load .eim Model"]
LISTEN["ποΈ Capture Audio"]
CLASSIFY["Run Classification"]
SCORE{"reggaeton > 95%?"}
OLED["π₯οΈ Display Score"]
LOG["π Write Log"]
ACTION["π‘ Trigger Test Routine"]
START --> UI
UI --> WAIT
WAIT --> LOAD
LOAD --> LISTEN
LISTEN --> CLASSIFY
CLASSIFY --> SCORE
SCORE -->|"No"| OLED
OLED --> LISTEN
SCORE -->|"Yes"| LOG
LOG --> ACTION
ACTION --> LISTEN
π Logging Every major operation is written to:
now = datetime.datetime.now()
dtFormatted = now.strftime(
"%Y-%m-%d %H:%M:%S"
)
Started
Listening
AI model ...
Firing threshold ...
Interrupted
This is useful for comparing inference behavior with the audio being played during experiments.
π Runtime Layout The code expects a slightly different runtime layout from the flat GitHub repository.
A practical deployment is:
/home/pi/reggaeton/
β
βββ reggaetonBeGone.py
βββ reggaetonbgone-linux-armv7-v4.eim
βββ whitrabt.ttf
βββ log.txt
β
βββ images/
βββ logo.png
This matters because the current source loads:
myPath + 'images/logo.png'
while the GitHub repository currently stores logo.png at repository root.
Either create the images/ directory as above or change the image path in the Python source.
π» Software Setup The application requires:
Python 3
Edge Impulse Linux SDK
PortAudio / PyAudio
Raspberry Pi GPIO
Pillow
Adafruit SSD1306
Adafruit GPIO
BlueZ
The original complete dependency procedure is preserved in:
π Installation git clone \
https://github.com/ronibandini/reggaetonBeGone.git
cd reggaetonBeGone
Prepare the runtime assets:
logo.png
whitrabt.ttf
Edge Impulse .eim model
myPath
selectedDeviceId
threshold
targetAddr
method
Run the application in a controlled test environment with your own Bluetooth speaker.
π Repository Structure reggaetonBeGone/
β
βββ reggaetonBeGone.py
βββ README.md
βββ LICENSE
β
βββ logo.png
βββ sticker.png
βββ whitrabt.ttf
βββ log.txt
π§ͺ Versions
Version 1.0 β February 2024 The version available in this GitHub repository.
Raspberry Pi 3
Edge Impulse audio classification
128Γ32 OLED
GPIO start button
Configured Bluetooth target
Two experimental test methods
Activity log
Version 2.0 β April 2024 Released to participants of the Nerdearla Chile 2024 workshop.
Improvements documented by the author include:
64-bit support
On-device Bluetooth scanning
Strike system to reduce false positives
Process cleanup
Updated ML model
Version 3.0 β November 2024 Created for the Ekoparty 2024 workshop.
The experimental release added further scanning, model and external-radio improvements.
π₯ Videos & Talks
Nerdearla Chile 2024 Complete Spanish workshop covering:
hardware
Machine Learning
assembly
limitations
Version 2.0
YouTube Workshop
Nerdearla Argentina 2025 The later talk follows the evolution from Reggaeton Be Gone to Pocket Gone .
π External References
π οΈ Hackster.io β Full Tutorial The original February 26, 2024 build documents the circuit, Raspberry Pi configuration, ML training, audio input, OLED interface and enclosure.
π¦π· Medium β Original Article Detailed development history, versions, ML workflow, results, conferences and press coverage.
π€ Public Model 16 kHz mono Edge Impulse model with reggaeton and otros labels.
π° Press
WIRED Published February 27, 2024.
Hackaday Published February 23, 2024.
Hackster News
Tom's Hardware Published February 28, 2024.
Euronews
Univision
La NaciΓ³n Feature on maker culture, Reggaeton Be Gone and other experimental machines.
π Related GitHub Projects
ποΈ Photon 2 Voice Command Audio keyword recognition with Edge Impulse running directly on a Particle Photon 2.
π Reading Time Audio TinyML on Arduino Nano 33 BLE Sense for recognizing the sound of paper page turns.
π¦ TI AM62A AI Traffic Light Computer Vision inference connected to a physical control system.
π Domestic ML LRAD Audio output triggered by Edge Impulse Computer Vision classification.
π Contracultura Maker Contracultura Maker is a book by Roni Bandini about maker culture, experimental electronics, AI, physical computing and technological autonomy.
Reggaeton Be Gone is one of the projects connected with this approach to building deliberately unusual technological artifacts.
π¬ Contact Roni Bandini
Maker Β· AI Developer Β· Writer
Buenos Aires, Argentina
Built with π΅ + Raspberry Pi + Edge Machine Learning + Bluetooth.
Ecosystem Role Standard MoltPulse indexed agent.
Show off your Pulse Score in your GitHub README to build trust and rank higher.
[](https://molt-pulse.com/agents/ronibandini/reggaetonBeGone)
Copy 
