Fernando / Sane.AI.Spectrogram v3 Public
Target: ECO Clone this project
Experiments EON Tuner
The EON Tuner helps you quickly run hyper-parameter sweeps that explore different pre-processing + model architectures optimized for your defined objectives. Clone this project to use the EON Tuner.

Target

Run #3

ECO

500 ms

200000 kB

200000 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

Run #9 (Nov 26 2025, 18:06:17, finished) Run #8 (Nov 26 2025, 17:56:45, cancelled) Run #7 (Nov 26 2025, 17:05:08, finished) Run #6 (Nov 26 2025, 17:00:51, cancelled) Run #5 (Nov 26 2025, 16:29:03, finished) Run #4 (Nov 21 2025, 20:43:47, finished) Run #3 (Nov 21 2025, 19:28:59, finished) Run #2 (Nov 21 2025, 19:00:10, finished) Run #1 (Nov 21 2025, 18:58:08, finished)
spectr-48d
PERFORMANCE
LATENCY
500 ms
RAM
200000 kB
ROM
200000 kB
Unused
TIME-SERIES INPUT

5000 ms |
500 ms |
Enabled

SPECTROGRAM
Parameter Value
id 38
show_axes true
dspBlockId 78
fft_length 128
frame_length 0.05
frame_stride 0.05
noise_floor_db -52
implementationVersion 4
KERAS
Parameter Value
id 39
dsp 38
layers [object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
batchSize 32
learningRate 0.005
trainTestSplit 0.2
trainingCycles 100
minimumConfidenceRating 0.6
ANOMALY-GMM
Parameter Value
id 75
axes audio RMS,audio Spectral Skewness,audio Spectral Kurtosis,audio Spectral Power 15.62 - 21.88 Hz,audio Spectral Power 21.88 - 28.12 Hz,audio Spectral Power 40.62 - 46.88 Hz,audio Spectral Power 171.88 - 178.12 Hz
clusterCount 4
minimumConfidenceRating 14.82

0

Help & documentation Get help on the forum Send feedback Contact sales
Join us
Terms of Service Privacy policy