Feature Extraction Techniques with Analysis of Confusing Words for Speech Recognition in the Hindi Language

Bhatt, Shobha; Jain, Anurag; Dev, Amita

doi:10.1007/s11277-021-08181-0

Cited by 19 publications

(3 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MFCC is the most popular feature extraction technique to the point where it is said to have become the standard feature extraction method for speech recognition [21]. Its popularity stems from the fact that it tries to mimic the human hearing system [40]. MFCC features are rooted in the recognized discrepancy of the human ear's critical bandwidths with frequency filters spaced linearly at low frequencies and logarithmically at high frequencies, which have been employed to keep the phonetically crucial aspects of the speech signal .…”

Section: Mel Frequency Cepstral Coefficient (Mfcc)mentioning

confidence: 99%

Challenges and Limitations in Speech Recognition Technology: A Critical Review of Speech Signal Processing Algorithms, Tools and Systems

Basak¹,

Agrawal²,

Jena³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

Speech recognition systems have become a unique human-computer interaction (HCI) family. Speech is one of the most naturally developed human abilities; speech signal processing opens up a transparent and hand-free computation experience. This paper aims to present a retrospective yet modern approach to the world of speech recognition systems. The development journey of ASR (Automatic Speech Recognition) has seen quite a few milestones and breakthrough technologies that have been highlighted in this paper. A step-by-step rundown of the fundamental stages in developing speech recognition systems has been presented, along with a brief discussion of various modern-day developments and applications in this domain. This review paper aims to summarize and provide a beginning point for those starting in the vast field of speech signal processing. Since speech recognition has a vast potential in various industries like telecommunication, emotion recognition, healthcare, etc., this review would be helpful to researchers who aim at exploring more applications that society can quickly adopt in future years of evolution. KEYWORDSSpeech recognition; automatic speech recognition (ASR); mel-frequency cepstral coefficients (MFCC); hidden Markov model (HMM); artificial neural network (ANN)

show abstract

Section: Mel Frequency Cepstral Coefficient (Mfcc)mentioning

confidence: 99%

Challenges and Limitations in Speech Recognition Technology: A Critical Review of Speech Signal Processing Algorithms, Tools and Systems

Basak¹,

Agrawal²,

Jena³

et al. 2023

Computer Modeling in Engineering &Amp; Sciences

View full text Add to dashboard Cite

show abstract

“…The study shows that some Hindi phonemes significantly impact on ASR, whereas some are confusing. Another comparative study of confusing Hindi words was performed to reduce the error in recognition of those words during building up an ASR system (Bhatt et al, 2021, Feature extraction techniques with analysis of confusing words for speech recognition in the Hindi language). The experiments showcased that depending on the error, different feature extraction techniques, classification techniques, acoustic model and pronunciation dictionaries can be selected to improve the performance of the ASR system.…”

Section: Review On Asr For Indian Languagesmentioning

confidence: 99%

A hierarchical automatic phoneme recognition model forHindi‐Devanagariconsonants using machine learning technique

Malakar

Keskar

Zadgaonkar³

2023

Expert Systems

View full text Add to dashboard Cite

A phoneme is perceptually the smallest distinct sound unit distinguished among words in a particular language. Every language has its own set of phonemes, and all the words are ordered sequences of phonemes. Therefore, phoneme recognition is essential to automatic speech recognition (ASR) systems. Phonemes of a language can be classified together using a single machine learning (ML) model through the direct classification (also known as the baseline or flat classification) approach. However, it is observed that the performance of such phoneme recognition degrades with the increase in the number of phoneme classes. The challenge is pronounced in languages with a larger number of phoneme classes, like Hindi, which has 48 phonemes.In this paper, we propose a speaker-independent hierarchical classification approach for 33 Hindi-Devanagari consonants/phonemes using cepstral features with ML techniques like support vector machine (SVM), random forest (RF) and fully connected deep neural network (DNN). In this hierarchical approach, a given phoneme is classified into successive subgroups until the particular phoneme class is identified.To perform the classification task, a binary or multi-class classifier is invoked for each internal (non-leaf) node in the hierarchy tree. Our model identified pairs of Optimal Feature Sets (based on mutual information) and the best suitable ML classifier for each internal decision node in the hierarchy using 10-fold cross-validation to help in efficient classification. Our proposed hierarchical model leads to better accuracy and 57% improved performance for phoneme recognition compared with the non-hierarchical, that is, the direct classification approach.

show abstract

“…De ahí que, una consecuencia de la baja CSE sea que se ha transformado en uno de los problemas fundamentales y sociales del estado peruano, los cuales causan trágicos efectos a largo plazo. Por otro lado, si el educador supiera las singularidades de cada estudiante, sería capaz de asistirlo para elaborar un verdadero entendimiento del aprendizaje; sin embargo, eso no se cumple en el contexto actual, razón por la cual, los estudiantes no sacian las tres exigencias psicológicas indispensables: independencia, idoneidad y comunicación (Petursdottir y Ragnarsdottir, 2019;Eguz, 2020;Putri et al, 2020;Arifin et al, 2021;Bani y Masruddin, 2021;Bhatt et al, 2021;Simanjuntak et al, 2022;Agustin et al, 2023), las cuales son el motivo para conseguir el desarrollo psicológico de la SE y otras impresiones buenas hacia su aprendizaje.…”

Section: Introductionunclassified

Quality of Educational Service and Student Satisfaction at P.E.I. Rosa de la Merced

Curo Aliaga,

Pilloca Moyna

2023

Reg Cient

View full text Add to dashboard Cite

Quality in education is a fundamental pillar to guarantee student satisfaction and, in turn, for the development of competencies and the realization of potentialities that reinforce the life project of the students. A study was carried out in the P.E.I Rosa de la Merced -Ate, 2022, to explore the relationship between educational service quality and student satisfaction. The analysis, of a basic type and with a quantitative approach, was oriented under a non-experimental design and a hypothetical-deductive method. The sample involved 105 students randomly selected from 168 third and fifth-level high school students. The results, with a p-value of 000 (less than α 0.05) and a Pearson's R test of 674, indicate a significant relationship between educational quality and student satisfaction. That is, the higher the educational service quality, the higher the level of student satisfaction, and vice versa.

show abstract

Feature Extraction Techniques with Analysis of Confusing Words for Speech Recognition in the Hindi Language

Cited by 19 publications

References 58 publications

Challenges and Limitations in Speech Recognition Technology: A Critical Review of Speech Signal Processing Algorithms, Tools and Systems

Challenges and Limitations in Speech Recognition Technology: A Critical Review of Speech Signal Processing Algorithms, Tools and Systems

A hierarchical automatic phoneme recognition model forHindi‐Devanagariconsonants using machine learning technique

Quality of Educational Service and Student Satisfaction at P.E.I. Rosa de la Merced

Contact Info

Product

Resources

About