Edge Impulse Experts / cough-monitor-audio (Created by Eivind Holt) Public
The EON Tuner helps you find the most optimal architecture for your embedded machine-learning application. Clone this project to use the EON Tuner.

Target

No name set

Arduino Nano 33 BLE Sense (Cortex-M4F 64MHz)

100 ms

256 kB

1024 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

98%
spectr-conv1d-8a0
PERFORMANCE
LATENCY
99 ms of 100 ms
RAM
24 kB of 256 kB
ROM
47 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTROGRAM

0.025 | 0.025 | -72

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 98%

Type Filters Kernel Rate
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.5

5/24/2022, 7:30:12 PM

95%
spectr-conv1d-bdb
PERFORMANCE
LATENCY
92 ms of 100 ms
RAM
24 kB of 256 kB
ROM
47 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTROGRAM

0.025 | 0.025 | -32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 95%

Type Filters Kernel Rate
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.25

5/24/2022, 7:30:29 PM

95%
spectr-conv1d-f6b
PERFORMANCE
LATENCY
211 ms of 100 ms
Exceeds target by 111 ms
RAM
38 kB of 256 kB
ROM
40 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTROGRAM

0.025 | 0.0125 | -72

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 95%

Type Filters Kernel Rate
Data augmentation
conv1d 16 3 -
conv1d 32 3 -
dropout - - 0.5

5/24/2022, 7:33:07 PM

94%
mfcc-conv1d-085
PERFORMANCE
LATENCY
329 ms of 100 ms
Exceeds target by 229 ms
RAM
24 kB of 256 kB
ROM
34 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFCC

0.05 | 0.025 | 40

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 94%

Type Filters Kernel Rate
conv1d 8 3 -
conv1d 16 3 -
dropout - - 0.5

5/24/2022, 7:31:17 PM

94%
mfcc-conv1d-d1d
PERFORMANCE
LATENCY
347 ms of 100 ms
Exceeds target by 247 ms
RAM
26 kB of 256 kB
ROM
44 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFCC

0.05 | 0.025 | 40

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 94%

Type Filters Kernel Rate
Data augmentation
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.25

5/24/2022, 7:31:43 PM

94%
mfe-conv1d-8d7
PERFORMANCE
LATENCY
102 ms of 100 ms
Exceeds target by 2 ms
RAM
20 kB of 256 kB
ROM
45 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.05 | 0.05 | 40

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 94%

Type Filters Kernel Rate
Data augmentation
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.5

5/24/2022, 7:31:34 PM

94%
mfe-conv1d-88c
PERFORMANCE
LATENCY
169 ms of 100 ms
Exceeds target by 69 ms
RAM
24 kB of 256 kB
ROM
35 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.05 | 0.025 | 40

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 94%

Type Filters Kernel Rate
Data augmentation
conv1d 8 3 -
conv1d 16 3 -
dropout - - 0.25

5/24/2022, 7:30:41 PM

93%
mfcc-conv1d-41c
PERFORMANCE
LATENCY
161 ms of 100 ms
Exceeds target by 61 ms
RAM
22 kB of 256 kB
ROM
70 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFCC

0.05 | 0.05 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 93%

Type Filters Kernel Rate
Data augmentation
conv1d 32 3 -
conv1d 64 3 -
conv1d 128 3 -
dropout - - 0.5

5/24/2022, 7:29:22 PM

93%
mfe-conv2d-998
PERFORMANCE
LATENCY
171 ms of 100 ms
Exceeds target by 71 ms
RAM
25 kB of 256 kB
ROM
36 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.05 | 0.05 | 40

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 93%

Type Filters Kernel Rate
Data augmentation
conv2d 8 3 -
conv2d 16 3 -
dropout - - 0.25

5/24/2022, 7:32:19 PM

92%
mfe-conv1d-a21
PERFORMANCE
LATENCY
179 ms of 100 ms
Exceeds target by 79 ms
RAM
24 kB of 256 kB
ROM
38 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.05 | 0.025 | 40

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 92%

Type Filters Kernel Rate
Data augmentation
conv1d 16 3 -
conv1d 32 3 -
dropout - - 0.25

5/24/2022, 7:29:29 PM

92%
mfe-conv2d-6d8
PERFORMANCE
LATENCY
97 ms of 100 ms
RAM
25 kB of 256 kB
ROM
36 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.05 | 0.05 | 40

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 92%

Type Filters Kernel Rate
Data augmentation
conv2d 8 3 -
conv2d 16 3 -
dropout - - 0.5

5/24/2022, 7:33:57 PM

89%
spectr-conv2d-628
PERFORMANCE
LATENCY
415 ms of 100 ms
Exceeds target by 315 ms
RAM
21 kB of 256 kB
ROM
36 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

SPECTROGRAM

0.075 | 0.075 | -72

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 89%

Type Filters Kernel Rate
Data augmentation
conv2d 8 3 -
conv2d 16 3 -
dropout - - 0.25

5/24/2022, 7:34:20 PM

70%
mfcc-conv1d-ad2
PERFORMANCE
LATENCY
252 ms of 100 ms
Exceeds target by 152 ms
RAM
21 kB of 256 kB
ROM
34 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFCC

0.05 | 0.025 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 70%

Type Filters Kernel Rate
conv1d 8 3 -
conv1d 16 3 -
dropout - - 0.25

5/24/2022, 7:29:10 PM

68%
mfcc-conv1d-1c1
PERFORMANCE
LATENCY
255 ms of 100 ms
Exceeds target by 155 ms
RAM
25 kB of 256 kB
ROM
45 kB of 1024 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFCC

0.05 | 0.025 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 68%

Type Filters Kernel Rate
Data augmentation
conv1d 8 3 -
conv1d 16 3 -
conv1d 32 3 -
conv1d 64 3 -
dropout - - 0.25

5/24/2022, 7:29:48 PM