[Question] Are input d-vectors for training assumed L2-normalized? about uis-rnn HOT 8 CLOSED

Yes, d-vectors are all assumed to have been L2 normalized.

And yes, you are correct, von Mises–Fisher distribution might be a better distribution here to use for UIS-RNN.

We just used Normal distribution as an approximation here due to its simplicity.

from uis-rnn.

thanks for replying so fast! :)

from uis-rnn.

I was struggling to overfit a single utterance segmentation without prior L2-normalization. I will let you know if now overfitting works as expected

from uis-rnn.

With L2-normalized speaker embeddings, given a single sequence to overfit (with only two speakers and some silence speaker), uis-rnn improves upon the initial accuracy (~35%) to ~55%, but does not reach higher accuracies. I'm using default transition_bias estimation and sigma2 initialization and adjustment.

from uis-rnn.

One more question: what is the semantics of "segments"? "nonoverlapping segments with max length of 400ms."?

from uis-rnn.

One more question: what is the semantics of "segments"? "nonoverlapping segments with max length of 400ms."?

Yes. Also, it really doesn't have to be 400ms. The length 400ms is what we found that works well on our dev / eval datasets.

from uis-rnn.

One more question about speaker embeddings. Are their values non-negative (i.e. ReLU is used prior to averaging / L2-normalization)?

I'm reimplementing a spectral clustering baseline from https://arxiv.org/abs/1710.10468 and https://github.com/wq2012/SpectralCluster, and depending on whether the values can be negative, the diffusion step may or may not be interpreted as random walk posterior probability after one step.

from uis-rnn.

They can be negative. We don't have ReLU after the last 256-dim linear layer.

from uis-rnn.