search

Snowdar / asv-subtools

An Open Source Tools for Speaker Recognition
Apache License 2.0
590 stars 135 forks source link

Can you write a tutorial of how to using backend to do score norm and fusion? #6

Closed zh794390558 closed 2 years ago

Snowdar commented 4 years ago

Yeah, thanks to your suggestion and I will write this in my free time (about this weekend).

Snowdar commented 4 years ago

Sorry, I am still busy now for this weekend. Sugget you to see subtools/recipe/voxceleb/gather_results_from_epochs.sh to know the example of score normalization. For score fusion, see subtools/weightScore.sh, subtools/fusionByLda.sh or subtools/fusitonBySvm.py to have a try. Note that, the format of trials is [target-id utt-id target/nontarget] and score is [target-id utt-id score].

zh794390558 commented 4 years ago

I will try and anytime is ok.

zh794390558 commented 4 years ago

using svm fusion, the result is minDCF is 1.0000 at threshold 1.0000 (p-target=0.05, c-miss=1,c-fa=1) how to debug this?

Snowdar commented 4 years ago

Hi, could you try the fusion with equal weights? Make sure it is no problem and then use LDA fusion to have a try. For SVM fusion, you should save the weights firstly and then apply these to your scores. By the way, you could also check the performance with EER metric.

On Aug 12, 2020, at 9:31 PM, Hui Zhang notifications@github.com wrote:

 using svm fusion, the result is minDCF is 1.0000 at threshold 1.0000 (p-target=0.05, c-miss=1,c-fa=1) how to debug this?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

zh794390558 commented 4 years ago

fusion with equal weights is wok, and I have not try LDA now, SVM fusion is not work for me.

Snowdar commented 4 years ago

Hi, are there negative weights? Try to plot a DET curve to observe the whole status.

Sent from my iPhone

On Aug 17, 2020, at 11:30 AM, Hui Zhang notifications@github.com wrote:

 fusion with equal weights is wok, and I have not try LDA now, SVM fusion is not work for me.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.

zh794390558 commented 4 years ago

why https://github.com/Snowdar/asv-subtools/blob/master/fusionBySvm.py#L142 txt_b is zero?

zh794390558 commented 4 years ago

does score norm always improve EER and minDCF?

Snowdar commented 4 years ago

Hi, it depends on the cohort set and you could not always select a good cohort.

On Sep 8, 2020, at 4:23 PM, Hui Zhang notifications@github.com wrote:

 does score norm always improve EER and minDCF?

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.