search

BrownsugarZeer / Multi_SSL

Combine sound source separation with SRP-PHAT to achieve multi-source localization.
MIT License
49 stars 12 forks source link

[FEAT] Can this project calculate the distance from the sound source to the microphone array #8

Open chimamaxianghahahahahaha opened 3 weeks ago

chimamaxianghahahahahaha commented 3 weeks ago

Checked other resources

Feature request

Can this project calculate the distance from the sound source to the microphone array

Motivation

Can this project calculate the distance from the sound source to the microphone array

Proposal (If applicable)

No response

BrownsugarZeer commented 3 weeks ago

Unfortunately it cannot.

letnnn commented 3 weeks ago

Is the inability to calculate distance a problem with this srp-phat algorithm itself or is it due to the limitations of the mini-microphone array?

BrownsugarZeer commented 3 weeks ago

SRP-PHAT itself can only get the direction where the sound come from.

letnnn commented 3 weeks ago

When I tried to measure the distance to the sound source with the TDOA I found that his time delay could not be calculated at all, I'm guessing that it's not too small a separation from my mike array? What do you think? Thanks for your answer.

BrownsugarZeer commented 2 weeks ago

Usually the spacing of most microphone arrays is sufficient and there won't be such a problem. Since the description is insufficient so I cannot give an feedback, sorry.

letnnn commented 2 weeks ago

Hi, here is the data related to my microphone array, 16KHz sampling rate, with such an array I am not accurate when calculating the latency results. Can you give me some suggestions? Thank you very much for your answer.

mics = torch.zeros((4, 3))
mics[0, :] = torch.Tensor([-0.02150, +0.03725, +0.059])
mics[1, :] = torch.Tensor([+0.02150, +0.03725, +0.059])
mics[2, :] = torch.Tensor([+0.02150, -0.03725, +0.059])
mics[3, :] = torch.Tensor([-0.02150, -0.03725, +0.059])
BrownsugarZeer commented 2 weeks ago

I don’t think there is anything wrong with this microphone array. Is it possible that there are uncertain factors in the experimental environment?