Automatic recognition of oral vowels in tone language: Experiments with fuzzy logic and neural network models

íkínbí, Lukman Adéwálé; Odéjobí, Odétúnjí íjídí

doi:10.1016/j.asoc.2010.04.018

Cited by 5 publications

(8 citation statements)

References 44 publications

(59 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Its involves generation of words from recorded speech (STT) [18] [19]. ASR is a technique aimed at converting speech signal into spoken word equivalents in text or action [20] in an accurate and efficient manner [21] as shown in figure 1. However, the conversion process should not depend on the speaker, acoustic signal or channel medium.…”

Section: Automatic Speech Recognition (Asr)mentioning

confidence: 99%

“…Most ASR performs poorly on intelligibility and naturalness due to non consideration of tonal cue in their design due to complexity in modeling tonal cue [13]. Also [20] observed that ASR for tone language involves complex task of simultaneously identifying tone and phoneme in speech signal.…”

Section: Research Progress In Sy Asrmentioning

confidence: 99%

“…In ASR research, tones have becomes an interesting attraction due to its: syllabic associativity -tones are associated with syllables which are building units of speech; unique fundamental frequency (F0) -in isolated utterance, each tone is characterized by a unique F0 curve [20] [27],and also the fact that in SY, tones are realized on vowels [20][30]. F0 is used to distinguish between lexical tones words that are otherwise phonemically identical [31].…”

Section: Research Progress In Sy Asrmentioning

confidence: 99%

“…Results from of both MLP and RNN reveal that both perform on training data than on test data with overall performance of RNN greater than MLP. Lack of speech technology for SY motivated the work by [20]. The paper asserted that recognition of vowels is a necessary prerequisite for utterances identification of tonal languages.…”

Section: Research Progress In Sy Asrmentioning

confidence: 99%

See 3 more Smart Citations

A review of Yorùbá Automatic Speech Recognition

Yusof

Atanda

Hariharan

2013

2013 IEEE 3rd International Conference on System Engineering and Technology

View full text Add to dashboard Cite

Automatic Speech Recognition (ASR) has recorded appreciable progress both in technology and application. Despite this progress, there still exist wide performance gap between human speech recognition (HSR) and ASR which has inhibited its full adoption in real life situation. A brief review of research progress on Yorùbá Automatic Speech Recognition (ASR) is presented in this paper focusing of variability as factor contributing to performance gap between HSR and ASR with a view of x-raying the advances recorded, major obstacles, and chart a way forward for development of ASR for Yorùbá that is comparable to those of other tone languages and of developed nations. This is done through extensive surveys of literatures on ASR with focus on Yorùbá. Though appreciable progress has been recorded in advancement of ASR in the developed world, reverse is the case for most of the developing nations especially those of Africa. Yorùbá like most of languages in Africa lacks both human and materials resources needed for the development of functional ASR system much less taking advantage of its potentials benefits. Results reveal that attaining an ultimate goal of ASR performance comparable to human level requires deep understanding of variability factors.

show abstract

Section: Automatic Speech Recognition (Asr)mentioning

confidence: 99%

Section: Research Progress In Sy Asrmentioning

confidence: 99%

Section: Research Progress In Sy Asrmentioning

confidence: 99%

Section: Research Progress In Sy Asrmentioning

confidence: 99%

See 2 more Smart Citations

A review of Yorùbá Automatic Speech Recognition

Yusof

Atanda

Hariharan

2013

2013 IEEE 3rd International Conference on System Engineering and Technology

View full text Add to dashboard Cite

show abstract

“…Therefore, in this paper, compared with some widely used classifiers, such as artificial neural networks (ANN) and support vector machine (SVM), the fuzzy expert system (FES) is employed as the main classifier. FES has been successfully used in many areas, such as speech recognition [12], power quality disturbances classification [13] and power system fault diagnosis [14].…”

Section: Introductionmentioning

confidence: 99%

A Smart Online Over-Voltage Monitoring and Identification System

et al. 2011

View full text Add to dashboard Cite

This paper proposes a complete and effective smart over-voltage monitoring and identification system. In recent years, smart grids are of the greatest interest in power system research. One of the main features of smart grid is their self-healing, which can continuously carry out online self-evaluation, discover existing faults, and correct them immediately. The over-voltage smart monitoring-identification-suppression systems play a key role in the construction of self-healing grids. In this paper, eight kinds of common over-voltage are discussed and analyzed. The S-transform algorithm is used to extract features of over-voltage. Aiming at the main features of each kind of over-voltage, six different characteristic quantities are proposed. A well designed fuzzy expert system and a support vector machine are employed as the classifiers to build a two-step identification model. The accuracy of the identification system is verified by field records. Results show that this system is feasible and promising for real applications.

show abstract