Table 6 Computational cost comparison
From: Dynamically localizing multiple speakers based on the time-frequency domain
| Â | # of parameters [million] | Average inference time of 1-s signal [seconds] |
|---|---|---|
CMS-DOA | 8.7 | 0.09 |
TF-DOAnet | 2.1 | 0.07 |
From: Dynamically localizing multiple speakers based on the time-frequency domain
| Â | # of parameters [million] | Average inference time of 1-s signal [seconds] |
|---|---|---|
CMS-DOA | 8.7 | 0.09 |
TF-DOAnet | 2.1 | 0.07 |