From: A recursive expectation-maximization algorithm for speaker tracking and separation
Algorithm | |||||||||
---|---|---|---|---|---|---|---|---|---|
Proposed | Baseline | ||||||||
ExperimentÔ | SIR(Input) | \(\textrm {SIR}_{1}(\hat {\mathbf {D}})\) | \(\textrm {SIR}_{1}(\hat {\mathbf {G}})\) | \(\textrm {SIR}_{2}(\hat {\mathbf {D}})\) | \(\textrm {SIR}_{2}(\hat {\mathbf {G}})\) | \(\textrm {SIR}_{1}(\hat {\mathbf {D}})\) | \(\textrm {SIR}_{1}(\hat {\mathbf {G}})\) | \(\textrm {SIR}_{2}(\hat {\mathbf {D}})\) | \(\textrm {SIR}_{2}(\hat {\mathbf {G}})\) |
#1 | 2.85 | 7.83 | 11.58 | 3.22 | 7.69 | 7.83 | 11.51 | 3.13 | 7.84 |
#2 | 2.12 | 4.92 | 5.83 | 3.24 | 5.17 | 4.17 | 4.41 | 1.53 | 3.29 |
#3 | 2.29 | 4.16 | 5.75 | 4.16 | 6.68 | 5.14 | 5.82 | 2.88 | 5.04 |
#4 | 5.02 | 6.99 | 8.54 | 3.60 | 4.50 | 6.38 | 8.42 | 2.97 | 3.49 |
#5 | 3.54 | 2.36 | 5.99 | 9.37 | 12.32 | 2.75 | 4.31 | 7.19 | 8.54 |
#6 | 4.03 | 6.74 | 8.79 | 3.42 | 5.43 | 3.18 | 4.33 | 6.13 | 6.80 |
#7 | 3.70 | 3.47 | 5.27 | 5.57 | 7.03 | 4.66 | 6.92 | 2.62 | 3.74 |
#8 | 2.86 | 2.85 | 4.47 | 6.41 | 8.14 | 2.56 | 4.03 | 5.92 | 7.85 |
#9 | 2.66 | 7.69 | 11.36 | 3.53 | 7.83 | 3.26 | 6.04 | 5.92 | 7.74 |
Mean | 2.66 | 7.69 | 11.36 | 3.53 | 7.83 | 3.26 | 6.04 | 5.92 | 7.74 |