Training corpus summary
projection of the training data. Each dot is one window of audio (explanation below). Hover a dot to hear the sound; click to open the recorded sample.
| Label | Samples | Fragments | Fragment durations0.5s bins · hover for counts | Windows | Window durations0.1s bins · hover for counts |
|---|---|---|---|---|---|
| bark | 0 | 0 | {b:0,0,0,0,0,0,0,0,0,0,0} | — | {b:0,0,0,0,0,0,0,0,0,0,0} |
| yap | 0 | 0 | {b:0,0,0,0,0,0,0,0,0,0,0} | — | {b:0,0,0,0,0,0,0,0,0,0,0} |
| background | 0 | 0 | {b:0,0,0,0,0,0,0,0,0,0,0} | — | {b:0,0,0,0,0,0,0,0,0,0,0} |
| wind | 0 | 0 | {b:0,0,0,0,0,0,0,0,0,0,0} | — | {b:0,0,0,0,0,0,0,0,0,0,0} |
| homestead | 0 | 0 | {b:0,0,0,0,0,0,0,0,0,0,0} | — | {b:0,0,0,0,0,0,0,0,0,0,0} |
| traffic | 0 | 0 | {b:0,0,0,0,0,0,0,0,0,0,0} | — | {b:0,0,0,0,0,0,0,0,0,0,0} |
| wildlife | 0 | 0 | {b:0,0,0,0,0,0,0,0,0,0,0} | — | {b:0,0,0,0,0,0,0,0,0,0,0} |
| gunshot | 0 | 0 | {b:0,0,0,0,0,0,0,0,0,0,0} | — | {b:0,0,0,0,0,0,0,0,0,0,0} |
| wrongdog | 0 | 0 | {b:0,0,0,0,0,0,0,0,0,0,0} | — | {b:0,0,0,0,0,0,0,0,0,0,0} |
Hover the table row above to highlight the dots in scatterplot. Click to select.
Each dot is a window. What are YAMNet windows and how are they made?
