Hindi Number Recognition using GMM

RaiGoyal, Himanshu; Koolagudi, Shashidhar G.

doi:10.5120/10589-5429

Cited by 1 publication

(1 citation statement)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This research, for the sake of distinguishing Brahman Tamil from Standard Tamil and other Tamil dialects, has referred to Brahman Tamil as 'Manipravalam' but has employed Tamil script instead of Grantha for the purpose of testing the output text. Goyal & Koolagudi (2013) have used the Mel-frequency cepstral coefficient (MFCC) technique in digit recognition for the Hindi language. They concluded that the performance of the system was dependent on a number of factors some of them being the effect of the speaker, age and gender.…”

Section: Introductionmentioning

confidence: 99%

Predicting user acceptance of Tamil speech to text by native Tamil Brahmans

Ramachandran¹

View full text Add to dashboard Cite

A thesis submitted in partial fulfilment of the requirement of Sheffield Hallam University for the degree of Doctor of Philosophy In the Cultural Communication and Computing Research Institute (C3RI) June 2018 Chapter 3: Literature review: This chapter presents a review of the literature on speech to text technology, various acceptance models, code-switching, language and politics, and on orthography. Chapter 4: Methods: This chapter deals with the research methods. It introduces the research philosophy that this research has adopted. The chapter discussess employment of software engineering as a design method and explains how the prototype was conceptualised, developed and evaluated. It also briefly deals with the sampling and pilot which is crucial to provide a useful insight on decisions taken on methods. Chapter 5: Evaluation of prototype: This chapter deals with the evaluation of the prototype and the rationale for some of the decisions in relation to prototype evaluation. Chapter 6: Proposed Mode: The proposed model to predict the user acceptance of speech to text in Tamil is introduced in this chapter. The proposed user acceptance model is one the main contributions of this research. Chapter 7: The studies: This chapter discusses the various studies and findings that were done in order to answer the main research question of predicting the user acceptance of speech to text in Tamil. It starts from the study of transliteration followed by a comparative study between two similar social groups which led to the observation of 'zha' pronunciation which in turn led to the target audience of this research-the Tamil Brahmans. Chapter 8: Discussion: The chapter discusses on the findings and studies along with relevant literature. The chapter also provides a reflection on the user acceptance model proposed in this thesis. The section on researcher's own experience is in the first person. Chapter 9: Personal reflection: This chapter provides a brief insight on researcher's own experiences, practices and observations as an insider. The chapter enables the reader to see through the researcher's lens. This chapter has used first person. Chapter 10: Conclusion: In addition to predicting the user acceptance, it provides an insight on the feasibility of such applications along with some suggestions that could be incorporated when designing and developing such applications. Finally, it enumerates the contribution to the knowledge and further work. * So what is your opinion/ experience, very specifically on pronunciation of the Brahmins and the non Brahmins? (clarifies: தமிழ் , தமிலழ பபாறுத் த வலரக் கும்

show abstract

Section: Introductionmentioning

confidence: 99%

Predicting user acceptance of Tamil speech to text by native Tamil Brahmans

Ramachandran¹

View full text Add to dashboard Cite

show abstract

Hindi Number Recognition using GMM

Cited by 1 publication

References 7 publications

Predicting user acceptance of Tamil speech to text by native Tamil Brahmans

Predicting user acceptance of Tamil speech to text by native Tamil Brahmans

Contact Info

Product

Resources

About