Nonlinear normalization of input patterns to speaker variability in speech recognition neural networks

Nejadgholi, Isar; Seyyedsalehi, Seyyed Ali

doi:10.1007/s00521-007-0151-5

Cited by 15 publications

(11 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The ratio of true distinctions in test data to all test data is introduced as accuracy of prediction. In the current study, comparison measurement was based on proper decrease of error and accuracy of prediction in test villages (12).…”

Section: Artificial Neural Network Modelingmentioning

confidence: 99%

Model of Cholera Forecasting Using Artificial Neural Network in Chabahar City, Iran

Pezeshki¹,

Tafazzoli-Shadpour²,

Nejadgholi³

et al. 2016

Int J Enteric Pathog

Self Cite

View full text Add to dashboard Cite

Background: Cholera as an endemic disease remains a health issue in Iran despite decrease in incidence. Since forecasting epidemic diseases provides appropriate preventive actions in disease spread, different forecasting methods including artificial neural networks have been developed to study parameters involved in incidence and spread of epidemic diseases such as cholera. Objectives: In this study, cholera in rural area of Chabahar, Iran was investigated to achieve a proper forecasting model. Materials and Methods: Data of cholera was gathered from 465 villages, of which 104 reported cholera during ten years period of study. Logistic regression modeling and correlate bivariate were used to determine risk factors and achieve possible predictive model onehidden-layer perception neural network with backpropagation training algorithm and the sigmoid activation function was trained and tested between the two groups of infected and non-infected villages after preprocessing. For determining validity of prediction, the ROC diagram was used. The study variables included climate conditions and geographical parameters. Results: After determining significant variables of cholera incidence, the described artificial neural network model was capable of forecasting cholera event among villages of test group with accuracy up to 80%. The highest accuracy was achieved when model was trained with variables that were significant in statistical analysis describing that the two methods confirm the result of each other. Conclusions: Application of artificial neural networking assists forecasting cholera for adopting protective measures. For a more accurate prediction, comprehensive information is required including data on hygienic, social and demographic parameters.

show abstract

Section: Artificial Neural Network Modelingmentioning

confidence: 99%

Model of Cholera Forecasting Using Artificial Neural Network in Chabahar City, Iran

Pezeshki¹,

Tafazzoli-Shadpour²,

Nejadgholi³

et al. 2016

Int J Enteric Pathog

Self Cite

View full text Add to dashboard Cite

show abstract

“…Moreover, this projection is similar to the tonotopic organization of the human peripheral auditory system [43]. Therefore, it has been used in many ANN structures for speech recognition in diverse researches [37,43,44].…”

Section: Modular Deep Neural Network (Mdnn)mentioning

confidence: 91%

“…Although indicated feature space transformations improve the recognition performance of DNNs, they are linear and cannot deal with nonlinear variations of the speech signal. In the nonlinear feature normalization method proposed in [37], a feed-forward neural network is first trained to map the input representations into both phonetic and speaker codes. Then, a training speaker with the highest phone accuracy is considered as the reference speaker.…”

Section: Previous Work On Using Dnns In Speech Recognitionmentioning

confidence: 99%

“…In this paper, a It has been shown that the robust performance of the human perception system relates to the special mode of signal processing in the brain. Some of the characteristics of this mode of processing are: (1) handling the information in both forward and backward passes, (2) the ability of analyzing linear and nonlinear components of its input, and (3) omitting the effect of each component if necessary [37,40,41]. According to this information, two speaker adaptation methods are proposed to approach a MDNN that can handle speaker variability.…”

Section: Introducing Mdnn For Speech Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

Toward growing modular deep neural networks for continuous speech recognition

Ansari

Seyyedsalehi

2016

Neural Comput & Applic

Self Cite

View full text Add to dashboard Cite

The performance drop of typical automatic speech recognition systems in real applications is related to their not properly designed structure and training procedure. In this article, a growing modular deep neural network (MDNN) for speech recognition is introduced. According to its structure, this network is pre-trained in a special manner. The ability of the MDNN to grow enables it to implement spatiotemporal information of the frame sequences at the input and their labels at the output layer at the same time. The trained network with such a double spatiotemporal (DST) structure has learned valid phonetic sequences subspace. Therefore, it can filter out invalid output sequences in its own structure. In order to improve the proposed network performance in speaker variations, two speaker adaptation methods are also presented in this work. In these adaptation methods, the network trains how to move distorted input representations nonlinearly to their optimal positions or to adapt itself based on the input information. To evaluate the proposed MDNN structure and its modified versions, two Persian speech datasets are used: FARSDAT and Large FARSDAT. As there is no frame-level transcription for large vocabulary speech datasets, a semi-supervised learning algorithm is explored to train MDNN on Large FARSDAT. Experimental results on FARSDAT verify that implementing the DST structure besides speaker adaptation methods achieves up to 7.3 and 10.6 % absolute phone accuracy rate improvement over the MDNN and typical hidden Markov model, respectively.Likewise, semi-supervised training of the grown MDNN on Large FARSDAT improves its recognition performance up to 5 %.

show abstract

“…This kind of neural network has also been successfully applied in clinical outcome prediction of myocardial infarction, mortality, surgical decision making on traumatic brain injury patients, recovery from surgery, pediatric, genecology, head trauma, and transplantation [7][8][9][10][11][12][13][14][15][16].…”

Section: Introductionmentioning

confidence: 99%

Nonlinear Poisson regression using neural networks: a simulation study

Fallah

Mohammad

et al. 2009

Neural Comput & Applic

Self Cite

View full text Add to dashboard Cite

We describe a novel extension of the Poisson regression model to be based on a multi-layer perceptron, a type of neural network. This relaxes the assumptions of the traditional Poisson regression model, while including it as a special case. In this paper, we describe neural network regression models with six different schemes and compare their performances in three simulated data sets, namely one linear and two nonlinear cases. From the simulation study it is found that the Poisson regression models work well when the linearity assumption is correct, but the neural network models can largely improve the prediction in nonlinear situations.

show abstract

Nonlinear normalization of input patterns to speaker variability in speech recognition neural networks

Cited by 15 publications

References 6 publications

Model of Cholera Forecasting Using Artificial Neural Network in Chabahar City, Iran

Model of Cholera Forecasting Using Artificial Neural Network in Chabahar City, Iran

Toward growing modular deep neural networks for continuous speech recognition

Nonlinear Poisson regression using neural networks: a simulation study

Contact Info

Product

Resources

About