Speech recognition of mandarin monosyllables

Li, Tze Fen

doi:10.1016/s0031-3203(03)00135-3

Cited by 8 publications

(4 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Multi-layer Perceptron [2][3][4], Hidden Markov Models [5][6][7], Recurrent Neural Network [8,9] and Dynamic Time Warping [10][11][12] are some common methods applied to recognize the speech signal.…”

Section: Introductionmentioning

confidence: 99%

Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

Seman

Bakar

2010

2010 International Conference on Science and Social Research (CSSR 2010)

View full text Add to dashboard Cite

Abstract-This paper describes speech signal modeling techniques which are suited to high performance and robust isolated word recognition. In this study, a speech recognition system is presented, specifically an isolated spoken Malay word recognizer which uses spontaneous and formally speeches collected from Parliament of Malaysia. Currently the vocabulary is limited to 25 words that can be pronounced exactly as it written and controls the distribution of the vocalic segments. The speech segmentation task is achieved by adopted energy based parameter and zero crossing rate measure with modification to better locates the beginning and ending points of speech from the spoken words. The training and recognition processes are realized by using Multi-layer Perceptron (MLP) Neural Networks with two-layer network configurations that are trained with stochastic error back-propagation to adjust its weights and biases after presentation of every training data. The Melfrequency Cepstral Coefficients (MFCCs) has been chosen as speech extraction approach from each segmented utterance as characteristic features for the word recognizer. Recognition results showed that the performance of the two-layer networks increased as the numbers of hidden neurons increased. The best network structures average classification rate is 84.731% with (150-25) configuration. Implementation results also showed that the conjugate gradient (CG) algorithm was more accurate and reliable than the Levenberg-Marquardt (LM) algorithm for the network complexities and data size considered in this study.

show abstract

Section: Introductionmentioning

confidence: 99%

Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

Seman

Bakar

2010

2010 International Conference on Science and Social Research (CSSR 2010)

View full text Add to dashboard Cite

show abstract

“…Wx(n, m) = E E a a 8(m _ ml + mk )e j(mi-mk)n (6) i=1lk= In the above sum item, as the situation of i = k, the sum item is independent item; as the situation of i . k, the sum item is cross item.…”

Section: Application Of Wigner Distribution In Feature Extractionmentioning

confidence: 99%

“…The feature parameters that we often used in speech recognition are Linear Prediction Cepstrum Coefficients (LPCC) [5], Mel-Frequency Cepstrum Coefficients (MFCC) [6], Perceptual Linear Predictive Cepstrum Coefficients (PLPC) [7], Difference Power Spectrum Cepstrum Coefficients (DPSCC) [8] and Perceptual Harmonic Cepstrum Coefficients (PHCC) [9], etc. This paper aims at improving the performance of speech recognition system in noisy environment, focuses on robust speech feature coefficients extraction and studies robust speech recognition system in noise.…”

Section: Introductionmentioning

confidence: 99%

Hybrid Model of Wigner Distribution/HMM and Self-organizing Neural Network for Speech Recognition

Chen

Wang

Lei

2006

The Proceedings of the Multiconference on "Computational Engineering in Systems Applications"

View full text Add to dashboard Cite

Based on the time-varying character of speech, this paper makes full use of excellent characters in Wigner Distribution (WD) and combines the homomorphic processing technique, proposes a new cepstrum coefficient, WD-MFCC. And apply this coefficient in a hybrid model of hidden Markov models and a self-organizing neural network for speech recognition in noisy environment. Experiments use three cepstrum coefficients (MFCC, DPSCC and WD-MFCC) into three speech recognition models respectively, results prove that with low SNR, compared with traditional Continuous Density HMM model (CDHMM with pure speech) and CDHMM-N model (CDHMM adding addictive noise), the hybrid model which introduced in this paper with WD-MFCC can obviously improve the recognition rate and the performance of speech recognition system in noisy environment.

show abstract

“…Though this limitation is not universal for any language (e.g., it is probably inappropriate for English), it was successfully applied to recognize French language with very-large vocabulary [20]. Another application of isolated syllable mode is Mandarin Chinese language with monosyllabic structure and thousands of syllables varied in tone [18]. This approach has been used in continuous ASR for Mandarin language with conventional HMM-GMM approach and syllable as a subword [17].…”

Section: Introductionmentioning

confidence: 99%

Towards the creation of reliable voice control system based on a fuzzy approach

Savchenko

2015

Pattern Recognition Letters

View full text Add to dashboard Cite

Speech recognition of mandarin monosyllables

Cited by 8 publications

References 56 publications

Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

Measuring the performance of isolated spoken Malay speech recognition using Multi-layer Neural Networks

Hybrid Model of Wigner Distribution/HMM and Self-organizing Neural Network for Speech Recognition

Towards the creation of reliable voice control system based on a fuzzy approach

Contact Info

Product

Resources

About