Integrating DNN–HMM Technique with Hierarchical Multi-layer Acoustic Model for Text-Dependent Speaker Verification

Laskar, Mohammad Azharuddin; Laskar, Rabul Hussain

doi:10.1007/s00034-019-01103-3

Cited by 12 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Laskar and Laskar 33 proposed integrating the deep neural network (DNN)‐hidden markov model technique with a hierarchical multilayer acoustic model for enhanced performance of the TD speaker authentication system. Using heirarchical multi‐layer acoustic model, the data alignment for DNN training is performed, which describes the speaker's idiosyncrasies more efficiently.…”

Section: Literature Reviewmentioning

confidence: 99%

Automatic speaker recognition from speech signal using bidirectional long‐short‐term memory recurrent neural network

Devi

Thongam

2020

Computational Intelligence

View full text Add to dashboard Cite

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Automatic speaker recognition from speech signal using bidirectional long‐short‐term memory recurrent neural network

Devi

Thongam

2020

Computational Intelligence

View full text Add to dashboard Cite

show abstract

“…It has been extensively used as feature extractors and also to build end‐to‐end systems. Besides, it has been integrated with classical methods to form hybrid techniques (Zhong, ; Larcher, ; Zeinali, , Laskar, ). The i‐vector‐DNN is one such hybrid technique.…”

Section: Introductionmentioning

confidence: 99%

A fuzzy‐clustering‐based hierarchical i‐vector/probabilistic linear discriminant analysis system for text‐dependent speaker verification

Laskar

2020

Expert Systems

Self Cite

View full text Add to dashboard Cite

In the i‐vector/probabilistic linear discriminant analysis (PLDA) technique, the PLDA backend classifier is modelled on i‐vectors. PLDA defines an i‐vector subspace that compensates the unwanted variability and helps to discriminate among speaker‐phrase pairs. The channel or session variability manifested in i‐vectors are known to be nonlinear in nature. PLDA training, however, assumes the variability to be linearly separable, thereby causing loss of important discriminating information. Besides, the i‐vector estimation, itself, is known to be poor in case of short utterances. This paper attempts to address these issues using a simple hierarchy‐based system. A modified fuzzy‐clustering technique is employed to divide the feature space into more characteristic feature subspaces using vocal source features. Thereafter, a separate i‐vector/PLDA model is trained for each of the subspaces. The sparser alignment owing to subspace‐specific universal background model and the relatively reduced dimensions of variability in individual subspaces help to train more effective i‐vector/PLDA models. Also, vocal source features are complementary to mel frequency cepstral coefficients, which are transformed into i‐vectors using mixture model technique. As a consequence, vocal source features and i‐vectors tend to have complementary information. Thus using vocal source features for classification in a hierarchy tree may help to differentiate some of the speaker‐phrase classes, which otherwise are not easily discriminable based on i‐vectors. The proposed technique has been validated on Part 1 of RSR2015 database, and it shows a relative equal error rate reduction of up to 37.41% with respect to the baseline i‐vector/PLDA system.

show abstract

Lhasa Dialect Recognition of Different Phonemes Based on TDNN Method

Khysru,

Qie,

Shi

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Integrating DNN–HMM Technique with Hierarchical Multi-layer Acoustic Model for Text-Dependent Speaker Verification

Cited by 12 publications

References 19 publications

Automatic speaker recognition from speech signal using bidirectional long‐short‐term memory recurrent neural network

Automatic speaker recognition from speech signal using bidirectional long‐short‐term memory recurrent neural network

A fuzzy‐clustering‐based hierarchical i‐vector/probabilistic linear discriminant analysis system for text‐dependent speaker verification

Lhasa Dialect Recognition of Different Phonemes Based on TDNN Method

Contact Info

Product

Resources

About