|
Sample #1 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |
Sample #2 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |
Sample #3 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |
Sample #4 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |
Sample #5 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |
Sample #6 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |
Sample #7 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |
Sample #8 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |
Sample #9 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |
Sample #10 | |||
---|---|---|---|
Ground-truth RIR | Wet (target) | Dry (source) | |
AudioLM-like (RIR) | RQ-Transformer-like (RIR) | VALL-E-like (RIR) | |
AudioLM-like (convolved) | RQ-Transf. (convolved) | VALL-E-like (convolved) | |
FVN (RIR) | Non-AR, discrete (RIR) | Non-AR, continuous (RIR) | |
FVN (convolved) | Non-AR, discrete (convolved) | Non-AR, cont. (convolved) |