• The Journal of the Acoustical Society of Korea
• /
• v.22 no.5
• /
• pp.360-366
• /
• 2003

Although multilayer perceptrons (MLPs) present several advantages against other pattern recognition methods, MLP-based speaker verification systems suffer from slow enrollment speed caused by many background speakers to achieve a low verification error. To solve this problem, the quantitative discriminative cohort speakers (QnDCS) method, by introducing the cohort speakers method into the systems, reduced the number of background speakers required to enroll speakers. Although the QnDCS achieved the goal to some extent, the improvement rate for the enrolling speed was still unsatisfactory. To improve the enrolling speed, this paper proposes the qualitative DCS (QlDCS) by introducing a qualitative criterion to select less background speakers. An experiment for both methods is conducted to use the speaker verification system based on MLPs and continuants, and speech database. The results of the experiment show that the proposed QlDCS method enrolls speakers in two times shorter time than the QnDCS does over the online error backpropagation(EBP) method.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.6
• /
• pp.585-591
• /
• 2002

While the multilayer perceptron(MLP) provides several advantages against the existing pattern recognition methods, it requires relatively long time in learning. This results in prolonging speaker enrollment time with a speaker verification system that uses the MLP as a classifier. This paper proposes a method that shortens the enrollment time through adopting the cohort speakers method used in the existing parametric systems and reducing the number of background speakers required to learn the MLP, and confirms the effect of the method by showing the result of an experiment that applies the method to a continuant and MLP-based speaker verification system.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.3
• /
• pp.88-98
• /
• 2002

The multilayer perceptron (MLP) has several advantages against other pattern recognition methods, and is expected to be used as the learning and recognizing speakers of speaker verification system. But because of the low learning speed of the error backpropagation (EBP) algorithm that is used for the MLP learning, the MLP learning requires considerable time. Because the speaker verification system must provide verification services just after a speaker's enrollment, it is required to solve the problem. So, this paper tries to make short of time required to enroll speakers with the MLP based speaker verification system, using the method of improving the EBP learning speed and the method of reducing background speakers which adopts the cohort speakers method from the existing speaker verification.

• Speech Sciences
• /
• v.9 no.2
• /
• pp.135-146
• /
• 2002

Generally a speaker verification system improves its system recognition ratio by regularizing log likelihood ratio, using a speaker model and its background speaker model that are required to be verified. The speaker-based cohort method is one of the methods that are widely used for selecting background speaker model. Recently, Gaussian-based cohort model has been suggested as a virtually synthesized cohort model, and unlike a speaker-based model, this is the method that chooses only the probability distributions close to basic speaker's probability distribution among the several neighboring speakers' probability distributions and thereby synthesizes a new virtual speaker model. It shows more excellent results than the existing speaker-based method. This study compared the existing speaker-based background speaker models and virtual speaker models and then constructed new virtual background speaker model groups which combined them in a certain ratio. For this, this study constructed a speaker verification system that uses GMM (Gaussin Mixture Model), and found that the suggested method of selecting virtual background speaker model shows more improved performance.

• Journal of Advanced Navigation Technology
• /
• v.5 no.1
• /
• pp.1-10
• /
• 2001

Speech has a good characteristics of which car drivers busy to concern with miscellaneous operation can make use in convenient handling and manipulating of devices. By utilizing this, this works proposes a speaker verification method for protecting cars from being stolen and identifying a person trying to access critical on-line services. In this, continuant phonemes recognition which uses language information of speech and MLP(mult-layer perceptron) which has some advantages against previous stochastic methods are adopted. The recognition method, though, involves huge computation amount for learning, so it is somewhat difficult to adopt this in speaker verification application in which speakers should enroll themselves at real time. To relieve this problem, this works presents a solution that introduces speaker cohort models from speaker verification score normalization technique established before, dividing background speakers into small cohorts in advance. As a result, this enables computation burden to be reduced through classifying the enrolling speaker into one of those cohorts and going through enrollment for only that cohort.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.571-573
• /
• 2004

Speaker verification systems based on multilayer perceptrons (MLPs) have good prospects in reliability and flexibility required as a successful authentication system. However, the poor learning speed of the error backpropagation (EBP) which is representative learning method of MLPs is the major defect to be complemented to achieve real-time user enrollments. In this paper, we implement an MLP-based speaker verification system and apply the existing two methods of the omitting patterns in instant learning (OIL) and the discriminative cohort speakers (DCS) to approach real-time enrollment. An evaluation of the system on a Korean speech database demonstrates the feasibility of the system as a speaker verification system of high performance.

• Asian-Australasian Journal of Animal Sciences
• /
• v.14 no.7
• /
• pp.1019-1023
• /
• 2001

An on-farm trial was conducted in temperature-controlled lactation rooms at a commercial pig farm to investigate the efficacy of broadcasting sow suckling grunts from day 4 of lactation, on increasing piglet growth to weaning. In the Broadcast treatment, sows and litters were exposed to a 3-min broadcast from loud-speakers every 42 min. The Control treatment was not exposed to the broadcast. All sows and litters had similar husbandry and piglets were provided with creep feed on the floor twice daily. In each of the three replicates in time, the Broadcast and Control treatments were allocated to different lactation rooms at random and there were 12 sows and litters per treatment per replicate. A total of four identical lactation rooms were available for the trial, each containing 28 conventional sow and litter crates with piglet heater in the creep area. A non-trial room separated the two treatment rooms in each replicate to minimise the chance that the broadcast grunt stimulation was audible to the Control treatment litters. Five "normal and average-looking" piglets from the trial litters were weighed twice, 7 d apart. The cohort of five piglets was identified by ear-tags and formed the experimental unit for the statistical analysis. The average (${\pm}SD$) age of piglets at initial weighing was 7.7(${\pm}2.22$) days. For each litter, mean piglet live weight at day 14 of lactation was estimated by linear regression of the two weights recorded seven days apart, when on average, the Broadcast treatment had been exposed to the stimulation for 10 days. Piglets in the Broadcast treatment were heavier (p<0.01) at day 14 of lactation compared to Control treatment (4.24 and 3.92 kg, respectively) and tended to have a greater average daily weight gain over the 7-d period (245 and 228 g/day, respectively; p<0.08). The results suggest piglet growth was improved by about 8% in response to the regular, timed broadcast of sow suckling grunts in the lactation shed. The independent contributions of milk and creep feed to the improved growth remain to be determined.

• Asian Pacific Journal of Cancer Prevention
• /
• v.17 no.2
• /
• pp.873-876
• /
• 2016

Background: Studies of quality of life (QoL) of oncological patients is carried out using questionnaires approved in many international clinical studies. The European Organization for Research and Treatment of Cancer EORTC QLQ-C30 (Quality of Life Questionnary-Core 30) and its special brain cancer module EORTC QLQ-BN20 are widely used in the world neurooncologic practice. They are available in more than 80 official versions of 30 languages of the world. Previously we used the official versions in Russian, which often causes difficulty in understanding for native Kazakh language speakers, who comprise more than 60% of our respondents. This was the reason for creating a version of Kazakh language. Therefore, in 2014 for the first time the process of adaptation of questionnaires to the Kazakh language was initiated. Materials and Methods: The translation process of questionnaires to Kazakh language was held in accordance with the requirements of the European Organization for Research and Treatment of Cancer EORTC on QoL and consisted of the following stages: preparation - translation - pilot testing - approval. The official permission of authors and "Guideline on translation" was obtained which was developed by the working group of the EORTC on QoL. The pilot testing of EORTC QLQ-C30 and QLQ-BN20 questionnaires was conducted on the basis of the Department of Central Nervous System Pathology of the "National Centre for Neurosurgery" in patients with malignant tumors of the central nervous system. Results: The official versions of the EORTC QLQ-C30 and QLQ-BN20 questionnaires in Kazakh language were introduced and adapted in practical neurosurgical operations in Kazakhstan. Conclusions: The approved versions of the questionnaires in Kazakh language are now available for mainstream use on the official website EORTC.com. The versions of these questionnaires can be used in domestic cohort studies and clinical practice in the Republic of Kazakhstan. The use of these tools for assessing QoL will help professionals in the planning of individual treatment strategies and selection of the necessary therapy.