devjonathan / Tutorial: Responding to your voice Public
Target: Himax WiseEye2 (Cortex-M55 400 MHz) + Ethos-U55-64 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

No name set

Himax WiseEye2 (Cortex-M55 400 MHz) + Ethos-U55-64

100 ms

2560 kB

16384 kB

Filters

Status

DSP type

Model type

View

Data set

Variant

Sort

General

F1-score

Precision

Recall

94%
mfe-conv2d-596
PERFORMANCE
LATENCY
7 ms of 100 ms
RAM
16 kB of 2560 kB
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.032 | 0.016 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 94%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.5

10/27/2022, 5:28:01 PM

93%
mfcc-conv2d-1be
PERFORMANCE
LATENCY
0 ms of 100 ms
RAM
Not available for this model
ROM
0 kB of 16384 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 | 93%

Type Filters Kernel Rate
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.5

10/27/2022, 5:43:13 PM

91%
mfe-conv2d-8c5
PERFORMANCE
LATENCY
4 ms of 100 ms
RAM
13 kB of 2560 kB
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.032 | 0.032 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 91%

Type Filters Kernel Rate
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.5

10/27/2022, 5:19:47 PM

90%
mfe-conv2d-39f
PERFORMANCE
LATENCY
0 ms of 100 ms
RAM
Not available for this model
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.05 | 0.05 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 90%

Type Filters Kernel Rate
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.25

10/27/2022, 5:40:24 PM

90%
mfe-conv2d-f96
PERFORMANCE
LATENCY
0 ms of 100 ms
RAM
Not available for this model
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 90%

Type Filters Kernel Rate
Data augmentation
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.5

10/27/2022, 5:51:05 PM

89%
mfcc-conv2d-9f6
PERFORMANCE
LATENCY
10 ms of 100 ms
RAM
19 kB of 2560 kB
ROM
0 kB of 16384 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 | 89%

Type Filters Kernel Rate
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.25

10/27/2022, 5:30:15 PM

89%
mfe-conv2d-c20
PERFORMANCE
LATENCY
3 ms of 100 ms
RAM
12 kB of 2560 kB
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.05 | 0.05 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 89%

Type Filters Kernel Rate
Data augmentation
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.25

10/27/2022, 5:35:36 PM

89%
mfcc-conv2d-156
PERFORMANCE
LATENCY
0 ms of 100 ms
RAM
Not available for this model
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFCC

0.02 | 0.01 | 40

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 89%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.5

10/27/2022, 5:47:16 PM

89%
mfcc-conv2d-206
PERFORMANCE
LATENCY
0 ms of 100 ms
RAM
Not available for this model
ROM
0 kB of 16384 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 | 89%

Type Filters Kernel Rate
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.25

10/27/2022, 5:15:32 PM

89%
mfcc-conv2d-efb
PERFORMANCE
LATENCY
8 ms of 100 ms
RAM
17 kB of 2560 kB
ROM
0 kB of 16384 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 | 89%

Type Filters Kernel Rate
Data augmentation
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.25

10/27/2022, 5:42:23 PM

89%
mfcc-conv2d-471
PERFORMANCE
LATENCY
8 ms of 100 ms
RAM
17 kB of 2560 kB
ROM
0 kB of 16384 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 | 89%

Type Filters Kernel Rate
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
conv2d 128 3 -
dropout - - 0.5

10/27/2022, 5:46:10 PM

89%
mfe-conv2d-8c4
PERFORMANCE
LATENCY
0 ms of 100 ms
RAM
Not available for this model
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.032 | 0.032 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 89%

Type Filters Kernel Rate
Data augmentation
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.25

10/27/2022, 5:48:32 PM

88%
mfcc-conv2d-82f
PERFORMANCE
LATENCY
6 ms of 100 ms
RAM
14 kB of 2560 kB
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFCC

0.032 | 0.032 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 88%

Type Filters Kernel Rate
Data augmentation
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
conv2d 128 3 -
dropout - - 0.25

10/27/2022, 5:26:27 PM

87%
mfcc-conv2d-a32
PERFORMANCE
LATENCY
19 ms of 100 ms
RAM
26 kB of 2560 kB
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFCC

0.02 | 0.01 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 87%

Type Filters Kernel Rate
conv2d 8 3 -
conv2d 16 3 -
conv2d 32 3 -
conv2d 64 3 -
dropout - - 0.25

10/27/2022, 5:26:51 PM

87%
mfe-conv2d-fd3
PERFORMANCE
LATENCY
5 ms of 100 ms
RAM
14 kB of 2560 kB
ROM
0 kB of 16384 kB
DSP NN Unused
TIME-SERIES INPUT

1000 ms |
1000 ms |
Enabled

MFE

0.02 | 0.02 | 32

ACCURACY (KERAS)
CLASSIFICATION (KERAS)

0.005 | 100 | 87%

Type Filters Kernel Rate
conv2d 16 3 -
conv2d 32 3 -
dropout - - 0.25

10/27/2022, 5:34:23 PM

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