Audio: Visual Demo
Next topic
Video
Notes
Code
Data
Examples per label:
2
4
8
Label
Audio samples
one
two
_background_noise_
Model
Base Tensorflow model
speech-commands
Model vocabulary
18w
FFT type
BROWSER_FFT
Training
Max Epochs:
10
20
40
100
200
Run
Epoch #:
Loss:
Prediction
Input:
test/one/01.png
test/two/01.png
test/two/02.png
test/other/left.png
test/other/three.png
Predict
Label
Probability
one
two