• Journal of Korea Society of Industrial Information Systems
• /
• v.13 no.5
• /
• pp.78-86
• /
• 2008

Many people have conveniently used various messengers to talk with remote friends or to send urgent files to remote co-workers. Recently, it is also possible to use messenger with user's image. However, these messenger technologies are of no use for the blind. In order to cope with this problem, we propose voice-based messenger with a Braille system for the blind. The proposed messenger enables the blind to listen to the received sentences from remote user. It also enables them to listen to the written sentences before sending to remote user for the purpose of checking that the sentences are correctly written. The Braille system for writing sentences can be implemented by using the programmable NXT system, which contains a 32-bit ARM-7 micro-controller, with 4 touch-sensors. Finally, we apply the Bluetooth technology for wireless communication between the Braille system and the proposed messenger.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.5B
• /
• pp.448-456
• /
• 2003

In this parer, an ABR (Available Bit Rate) congestion control algorithm for voice transmission in ATM networks was proposed. To deal with the network congestion problem, not only the buffer level of a switch but also the variation of the buffer level were considered. Also, to resolve the unfairness among sources where the bit transfer rates vary, a loading factor that is used to adjust the bit rate was introduced. To show the superiority of this paper over others, simulation was done with a network of 7 voice sources and 4 switches, which was represented by Petri net model. ExSpect was used for simulation. The simulation results showed that there was improvement in network utilization and that unfairness among sources were resolved a lot.

• The KIPS Transactions:PartC
• /
• v.12C no.2 s.98
• /
• pp.251-260
• /
• 2005

On the internet phone or PSTN(Public Switched Telephone Network), the automatic callback is an useful service in the case of busy state when one user calls the other. By using this service, automatic redial is possible when the other party hangs up. However, in the basic automatic callback service, the user who wants callback should wait until the other party hangs up even in the case of emergency. Therefore in this paper, to solve this problem we have extended CPL(Call Processing Language) and, within user system we have included and linked this extended CPL processing module and dialog event package which processes SIP INVITE initiated dialog state informations. We have implemented this system for being used in SIP(Session Initiation Protocol)-based VoIP(Voice over IP) system.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.12 no.9
• /
• pp.4119-4124
• /
• 2011

In a MPC coding using excitation source of voiced and unvoiced, it would be a distortion of speech waveform in case of increasing or decreasing of speech signal amplitude in a frame. This is caused by normalization of synthesis speech signal in the process of restoration the multi-pulses of representation section. To solve this problem, this paper present a method of amplitude compensation(AC-MPC) in a multi-pulses each pitch interval in order to reduce distortion of speech waveform. I was confirmed that the method can be synthesized close to the original speech waveform. And I evaluate the MPC and AC-MPC using amplitude compensation method. As a result, SNRseg of AC-MPC was improved 0.7dB for female voice and 0.7dB for male voice respectively. Compared to the MPC, SNRseg of AC-MPC has been improved that I was able to control the distortion of the speech waveform finally. And so, I expect to be able to this method for cellular phone and smart phone using excitation source of low bit rate.

• Journal of IKEEE
• /
• v.28 no.2
• /
• pp.168-173
• /
• 2024

Speaker recognition refers to a technology that analyzes voice frequencies that are different for each individual and compares them with pre-stored voices to determine the identity of the person. Deep learning-based speaker recognition is being applied to many fields, and pet robots are one of them. However, the hardware performance of pet robots is very limited in terms of the large memory space and calculations of deep learning technology. This is an important problem that pet robots must solve in real-time interaction with users. Lightening deep learning models has become an important way to solve the above problems, and a lot of research is being done recently. In this paper, we describe the results of research on lightweight speaker recognition for pet robots by constructing a voice data set for pet robots, which is a specific command type, and comparing the results of models using residuals. In the conclusion, we present the results of the proposed method and Future research plans are described.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2012.05a
• /
• pp.213-215
• /
• 2012

The current Navigation System for the Visually Impaired Person has a short and limited communication distance and can't receive enough information from Visually Impaired Person to assist directly. In addition, because the path is dangerous and incomplete for the Visually Impaired Person, moving with White Stick is still inconvenient and dangerous. To solve this problem we implement communication that can send and receive video, voice, location information between the Visually Impaired Person's Android platform and assistant's PC.

• The Journal of the Acoustical Society of Korea
• /
• v.17 no.3E
• /
• pp.12-21
• /
• 1998

In this paper, we describe the one channel five-way, V/U/M/N/S (Voice/Unvoice/Nasal/Silent), classification algorithm for automatically classifying speech. The decision making process is viewed as a pattern viewed as a pattern recognition problem. Two aspects of the algorithm are developed: feature selection and classifier type. The feature selection procedure is studied for identifying a set of features to make V/U/M/N/S classification. The classifiers used are a vector quantization (VQ), a neural network(NN), and a decision tree method. Actual five sentences spoken by six speakers, three male and three female, are tested with proposed classifiers. From a set of measurement tests, the proposed classifiers show fairly good accuracy for V/U/M/N/S decision.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2007.11a
• /
• pp.325-329
• /
• 2007

The mobile and multimedia service on wireless network have been leaded from the improvement of telecommunication techniques. A typical multimedia service, a video transmission usually requires larger bandwidth than voice transmission. Many channel allocation algorithms for Orthogonal Frequency Division Multiple Access (OFBMA) to use resources more efficiently, Previous channel allocation algorithms have developed with an assumption that the data traffic is constant bit rate (CBR). However, existing algorithms are not suitable to video traffic because it usually generates a variable bit rate (VBR) traffic. In this paper, we proposed a new channel allocation algorithm called a queue-based channel allocation. it is more suitable to transmit the video traffics. Also, a problem are notified in case of realtime generated video traffic and a corresponding heuristic solution was proposed.

• Korean Management Science Review
• /
• v.25 no.2
• /
• pp.25-42
• /
• 2008

CDMA/TOD with asymmetric capacity allocation between uplink and downlink is a highly attractive solution to support the next generation mobile systems. This is because flexible asymmetric allocation of capacity to uplink and downlink usually improves the utilization of the limited bandwidth. In this paper, we mathematically formulate an optimal timeslot and code allocation problem, which is to maximize the total utility considering the numbers of codes(channels) allocated to each data class and the forced terminations of previously allocated codes. We also suggest a real-time integrated timeslot and code allocation scheme using Lagrangean relaxation and subgradient optimization techniques. Experimental results show that the proposed scheme provides high-quality solutions in a fast time.

• The Journal of the Acoustical Society of Korea
• /
• v.22 no.3E
• /
• pp.124-132
• /
• 2003

We propose a new method for speech/non-speech classifiers based on concepts of the adaptive boosting (AdaBoost) algorithm in order to detect speech for robust speech recognition. The method uses a combination of simple base classifiers through the AdaBoost algorithm and a set of optimized speech features combined with spectral subtraction. The key benefits of this method are the simple implementation, low computational complexity and the avoidance of the over-fitting problem. We checked the validity of the method by comparing its performance with the speech/non-speech classifier used in a standard voice activity detector. For speech recognition purpose, additional performance improvements were achieved by the adoption of new features including speech band energies and MFCC-based spectral distortion. For the same false alarm rate, the method reduced 20-50% of miss errors.