Rizwan Ishaq

16 Questions 18 Answers 0 Followers

Questions related from Rizwan Ishaq

Which acoustic features are better for language recognition task?

The language recognition uses the Shift Delta Coefficients(SDC) as acoustic features. Some papers uses only SDC(i.e. 49 for each frame), while some uses MFCC(c0-c6)+SDC (total of 56 for each...

17 March 2017 5,691 3 View

Opus Palyload in RTP packets?

I have raw opus payload from RTP packets, What I am doing is as: I extracted first 12 bytes as the RTP Header, and rest as the OPUS payload, I want ot know, how can I use this RAW OPUS payload to...

17 June 2016 7,477 1 View

Speaker Recognition PLDA training?

If we train our PLDA with microphone data only, and test with Phone data, will it effects the system performance? and If we train with large amount of data of microphone and with less data of...

06 June 2016 1,354 1 View

Influence of mismatch speech utterance length between enrollment and training hyperparameters on Speaker Recognition performance?

What would be the effect of the speech utterance length on speaker recognition. i.e if T, UBM, LDA, PLDA-----> are trained on short utterance i.e. from 3 to 15 seconds, but enrollment of...

24 May 2016 5,060 1 View

Is there any way to train the GMM for very very big data, I tried with Python sklearn gmm, but it give memory error?

Any efficient way of GMM estimation for very very big data???

04 May 2016 2,913 3 View

What would be better features MFCC or MFCC+delta+deltadelta in speaker recogntion?

What are the better features for speaker recognition MFCC or MFCC+delta+deltadelta? I tested my system with both type of features, and I got better results with mfcc only? Any comments about these...

02 May 2016 2,013 6 View

In UBM-GMM based speaker recognition, how much data is enough for UBM?

In UBM-GMM base speaker recognition, we need background GMM (UBM), how much data is enough, and what about adapting variances and weights to the desired speaker?? Or is it enough to adapat mean?

25 April 2016 8,801 1 View

Speaker/Langugage Recognition step?

I recently started to work on Speaker/Language recogntion using i-vector, and after consluting with researcher on researchgate, I came to the following steps:1) Databasei) Developement dataset...

21 April 2016 4,161 8 View

Is there any efficients and faster Gaussian Mixture Model (GMM) implementation, as Matlab/Pyhton(sklearn) are not so faster?

Is there any faster GMM implementation available, for Python or Matlab?

14 April 2016 7,134 4 View

If we don't have labeled developement data, how we train LDA/PLDA in speaker recognition?

In speaker recognition, to train LDA and PLDA we need labeled developement data, if we don't have labeled developement data for training LDA and PLDA, how we train the LDA and PLDA, and which...

14 April 2016 8,167 1 View

Can I have more than 1 speaker speech utterance in Developement data of the speaker recognition?

For speaker recognition, we need developement data to train T, UBM. Is it possible that the single speech sample have more than 1 speaker i.e in s = s1+s2+s3 ??

13 April 2016 3,565 3 View

Can anybody suggest recent language detection paper using speech signal?

Any recent state of the art review paper about language detection?

29 March 2016 7,378 2 View

How many factors are needed for total variable space (T) in speaker recognition, assum we have 24 speaker?

I tried to estimate total variable space (T) for i-vector estimation, in speaker recognition, but if i select the factor number to more than 100 it gave wrong results but with 100 or less it...

21 March 2016 2,840 1 View

What is the optimal time for training/testing the UBM-GMM and i-vector based system?

Two System (Speaker Recognition) UBM-GMM Optimal time for training and testing the system i-vector Optimal time for training and testing the system

01 March 2016 277 4 View

How much accurate UBM-GMM based speaker Recognition is?

I am currently working on Speaker recognition and implement the UBM-GMM based speaker recognition system, and test on the clean data? I just want to know how much better this system is in...

23 February 2016 7,956 9 View

Can any one suggest the stopping creteria for EM-GMM convergence for speaker recognition application?

I am trying to write the GMM-EM for speaker recognition, I need to know what is the stopping or convergence criteria for EM?

18 February 2016 2,528 2 View