Efficient Vector Code-book Generation using K-means and Linde-Buzo-Gray (LBG) Algorithm for Bengali Voice Recognition

Syfullah, Sadi Mohammud; Zakaria, Zareen Binte; Uddin, Md. Palash; Rabbi, Md. Fazle; Afjal, Masud Ibn; Nitu, Adiba Mahjabin

doi:10.1109/icaeee.2018.8642994

Cited by 6 publications

(6 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The volume of the datasets used in this work is comparable to other similar researches, such as [18], [31], and [32]. Particularly for Bengali voice recognition, Syfullah et al.…”

Section: Methodssupporting

confidence: 63%

“…Therefore, each dataset consists of 40 × 7 = 280 samples, and considering two datasets (vowels and words), we have 560 samples in total. Figure 2 The volume of the datasets used in this work is comparable to other similar researches, such as [18], [31], and [32]. Particularly for Bengali voice recognition, Syfullah et al [18] used only 20 different sample inputs for each of their Bengali characters to classify.…”

Section: Speech Collection and Dataset Preparationsupporting

confidence: 61%

“…Figure 2 The volume of the datasets used in this work is comparable to other similar researches, such as [18], [31], and [32]. Particularly for Bengali voice recognition, Syfullah et al [18] used only 20 different sample inputs for each of their Bengali characters to classify. In a neural network-based word classification, Selvan and Rajesh [31] used 42 samples for each of their words.…”

Section: Speech Collection and Dataset Preparationsupporting

confidence: 61%

“…Particularly for Bengali voice recognition, Syfullah et al. [18] used only 20 different sample inputs for each of their Bengali characters to classify. In a neural network‐based word classification, Selvan and Rajesh [31] used 42 samples for each of their words.…”

Section: Methodsmentioning

confidence: 99%

“…They found that MFCCbased classification outperforms raw audio-based classification. The authors of [18] performed recognition of Bengali speech characters (vowels and consonants) using MFCC features. On top of these, Badhon et al [19] remarked some state-of-the-art studies in Bengali Speech Recognition up to the year of 2019.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

How many Mel‐frequency cepstral coefficients to be utilized in speech recognition? A study with the Bengali language

Hasan

Hossain

2021

The Journal of Engineering

View full text Add to dashboard Cite

Speech‐related research has a wide range of applications. Most speech‐related researches employ Mel‐frequency cepstral coefficients (MFCCs) as acoustic features. However, finding the optimum number of MFCCs is an active research question. MFCC‐based speech classification was performed for both vowels and words in the Bengali language. As for the classification model, deep neural network (DNN) with Adam optimizer was used. The performances were measured with five different performance metrics, namely confusion matrix, classification accuracy, area under curve of receiver operating characteristic (AUC‐ROC), F1 score, and Cohen's Kappa with four‐fold cross‐validations at different number of MFCCs. All performance metrics gave the best score for 24/25 MFCCs; hence it is suggested that the optimum number of MFCCs should be 25, although many existing studies use only 13 MFCCs. Furthermore, it is verified that increasing the number of MFCCs yields better classification metrics with lower computational burden than the increment of hidden layers. Lastly, the optimum number of MFCCs obtained from this study was used in a more improved DNN model, from which 99% and 90% accuracies were achieved for vowel and word classification, respectively, and the vowel classification score outperformed state‐of‐the‐art results.

show abstract

“…The volume of the datasets used in this work is comparable to other similar researches, such as [18], [31], and [32]. Particularly for Bengali voice recognition, Syfullah et al.…”

Section: Methodssupporting

confidence: 63%