Table 2
Speech source localization performance in hit rate (Eq. (16)) for the scenarios with a vacuum cleaner noise as localized interferer. Obtained with all different localization approaches – spatial probability only (DOA, Eq. (14)), speech quality measure only (ASQM, Eq. (15)), and the combination of DOA and ASQM (DOA × ASQM, Eq. (13)) – and ASR-based speech quality measures: M-Measure (MM), Matched Phoneme (MaP) filters, and Inverse entropy (iEnt). Results are shown for the anechoic (top panel) and the reverberant office (bottom panel) environment depending on the SNR, the best performing approach per SNR is highlighted in bold. The SP column denotes the hit rate with respect to the target speech source. In addition, the percentage of estimates that considered the interferer as target is given in the column labeled VC.
| Anechoic |
||||||||
|---|---|---|---|---|---|---|---|---|
| DOA | ASQM | DOA × ASQM | ||||||
| SNR | SP | VC | MM | MaP | iEnt | MM | MaP | iEnt |
| −10 | 0 | 100 | 1 | 60 | 3 | 14 | 0 | 0 |
| −5 | 0 | 100 | 0 | 69 | 1 | 86 | 2 | 0 |
| 0 | 0 | 100 | 0 | 75 | 1 | 98 | 23 | 22 |
| 5 | 0 | 100 | 0 | 66 | 0 | 100 | 65 | 76 |
| 10 | 0 | 100 | 1 | 51 | 1 | 100 | 91 | 95 |
| 15 | 7 | 93 | 3 | 48 | 0 | 100 | 98 | 99 |
| 20 | 56 | 44 | 4 | 49 | 0 | 100 | 100 | 100 |
Office |
||||||||
| DOA | ASQM | DOA × ASQM | ||||||
| SNR | SP | VC | MM | MaP | iEnt | MM | MaP | iEnt |
| −10 | 0 | 100 | 0 | 24 | 0 | 0 | 0 | 0 |
| −5 | 0 | 100 | 7 | 50 | 1 | 0 | 1 | 0 |
| 0 | 0 | 100 | 6 | 57 | 1 | 24 | 14 | 0 |
| 5 | 0 | 100 | 9 | 60 | 0 | 77 | 35 | 2 |
| 10 | 0 | 100 | 6 | 64 | 0 | 96 | 75 | 69 |
| 15 | 9 | 91 | 13 | 79 | 0 | 100 | 96 | 93 |
| 20 | 58 | 42 | 22 | 76 | 0 | 100 | 100 | 92 |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.