• Proceedings of the Korea Contents Association Conference
• /
• 2006.11a
• /
• pp.75-79
• /
• 2006

In a VoIP(Voice over IP) system which support the realtime speech service, speech quality is deteriorated by the delay, the jitter, the loss, and the reversed packet order. In this thesis, APBC for receiver site is proposed, which compensate the jitter by the adaptive playout algorithm and conceal the packet loss, and align the packet order. Also, a VoIP application system is implemented, and the performance of APBC is verified on the implemented system by measuring the processing speed and the speech quality. From the result, processing speed is 257$\mu$sec, which is fast enough to deal with packet being received in realtime. Also, the speech quality by MOS(Mean Opinion Score) is improved as 18 percent compared with algorithm of fixed playout delay.

• Speech Sciences
• /
• v.11 no.2
• /
• pp.181-192
• /
• 2004

This paper presents an efficient speech interactive agent rendering smooth car navigation and Telematics services, by employing embedded automatic speech recognition (ASR), distributed speech recognition (DSR) and text-to-speech (ITS) modules, all while enabling safe driving. A speech interactive agent is essentially a conversational tool providing command and control functions to drivers such' as enabling navigation task, audio/video manipulation, and E-commerce services through natural voice/response interactions between user and interface. While the benefits of automatic speech recognition and speech synthesizer have become well known, involved hardware resources are often limited and internal communication protocols are complex to achieve real time responses. As a result, performance degradation always exists in the embedded H/W system. To implement the speech interactive agent to accommodate the demands of user commands in real time, we propose to optimize the hardware dependent architectural codes for speed-up. In particular, we propose to provide a composite solution through memory reconfiguration and efficient arithmetic operation conversion, as well as invoking an effective out-of-vocabulary rejection algorithm, all made suitable for system operation under limited resources.

• Journal of the Ergonomics Society of Korea
• /
• v.18 no.3
• /
• pp.105-125
• /
• 1999

As speech is the human being's most natural communication medium, using it gives many advantages. Currently, most user interfaces of a computer are using a mouse/keyboard type but the interface using speech recognition is expected to replace them or at least be used as a tool for supporting it. Despite the advantages, the speech recognition interface is not that popular because of technical difficulties such as recognition accuracy and slow response time to name a few. Nevertheless, it is important to optimize the human-computer system performance by improving the usability. This paper presents a set of guidelines for designing speech recognition interfaces and provides a method for evaluating the usability. A total of 113 guidelines are suggested to improve the usability of speech-recognition interfaces. The evaluation method consists of four major procedures: user interface evaluation; function evaluation; vocabulary estimation; and recognition speed/accuracy evaluation. Each procedure is described along with proper techniques for efficient evaluation.

• Speech Sciences
• /
• v.1
• /
• pp.103-111
• /
• 1997

The purpose of this study was to investigate the effect of the speaking rate on the duration of syllable before boundary. The materials used were four types of syllable-boundary sequences(Go-'Ga' Boundary-Gu) in a paragraph. The duration of 'Ga' syllables before 4 level of boundary was measured, and all of the measurements were taken from signals and spectrograms made by the $Signalyze^{TM}$ 3.04 for Power Mac 7200. Subjects were six female speakers who read the materials at fast, normal, and slow speed five times. The results show that (1) the slower the speaking rate becomes, the longer the duration of syllable before boundary, (2) the duration rank of syllable before each boundary does not correspond to the level of boundary, eg. at fast speed, = < #, + < $ ; at normal speed, +, #, = < $ ; at slow speed, + < =, #, $, and (3) the syllable before sentence boundary is less influenced than syllable before another boundary.

• Journal of Korea Society of Industrial Information Systems
• /
• v.7 no.2
• /
• pp.68-73
• /
• 2002

Many paper about the adaptive algorithm based to LS(Least Square) to improve convergence speed are already presented. In this paper, we propose a wavelet based adaptive algorithm which improves the convergence speed and reduces computational complexity, and adapt two kinds of adaptive noise cancelers using the characteristic of speech signal. We compared the performance of the nosed algorithm with time and frequency domain adaptive algorithm using computer simulation of adaptive noise canceler based on synthesis speech. As the result the proposed algorithm is suitable for adaptive signal processing area using speech or acoustic signal.

• MALSORI
• /
• no.44
• /
• pp.93-104
• /
• 2002

There are many speech analysis tools available. Among them real-time analysis tool is very useful for interactive experiments. A real-time speech analysis tool was implemented using Matlab. Matlab is a very widely used general purpose signal processing tool. In general, its computational speed is relatively lower than that of the codes from conventional programming languages. Especially, real-time analysis including input of signal and output of the result was not possible in the past. However, due to the improvement of computing power of PCs and inclusion of real-time I/O toolboxes in Matlab, real-time analysis is now possible in some extent by Matlab only. In this experiment, we tried to implement a real-time speech analysis tool using Matlab. Pitch and spectral information is computed in real-time. From the result it is shown that such real-time applications can be implemented easily using Matlab.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.3
• /
• pp.293-303
• /
• 2002

This paper implements optimized Time Scale Modification (TSM) system using ITU G.729 speech decoder and Dual SOLA algorithm. The proposed system assume 8 Kz sampling rate, 80 samples/frame input speech from the ITU G.729 speech Decoder and the TSM (Time Scale Modification) feature of Dual SOLA produces the high quality output speech that was slow-down or speed up as a user's choice. Especially, the proposed Optimized Dual SOLA base on various simulations and theoretical analysis, and the additional interpolation procedure of the speech makes it possible to setup high performance integrated TSM system at the maximum time scale modification rate. The system performance is analyzed and verified with various input speech and playback speed.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.8 no.2
• /
• pp.139-144
• /
• 2014

Speech discrimination is affected by the speed of speech. The speed of speech can be adjusted at the pause duration, the pause duration can take the resting time to avoid in overloading information. The study will be examine the effects of aging and audiological rehabilitation, and the auditory processing as basic research to investigate the normative data. 7 boys and 8 girls were participated. They have no problem with speech language pathologically and audiologically. There are 4 sets of test implement, and each test set was made out with 20 3-syllable words. Pause duration of all of these words are adjusted in normal(250 ms), slow(500 ms) and very slow(1000 ms). There are 4 words for a multiple-choice that including one word with written correctly and three words with written 1 phoneme wrong. Participant hear the word, and then have to choose one. Speech discrimination score in 250, 500, 1,000 ms of pause duration were $73{\pm}19.4%$, $84{\pm}12.2%$, $88{\pm}8.8%$, respectively.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2014.04a
• /
• pp.358-359
• /
• 2014

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2014.04a
• /
• pp.360-361
• /
• 2014