• 지식경영연구
• /
• 제19권1호
• /
• pp.147-174
• /
• 2018

Recently, firms have been putting forth significant efforts to fulfill various demands and high expectations of customers. The role and importance of customer centers as a direct contact point for customer relationship management are more emphasized than previously. A customer center draws attention as a new alternative to secure corporate competitiveness as it contributes to sales increase, being in a position to satisfy customers' needs by ensuring customers' access to information. A customer center is an aggregation of various information and communication technologies. In particular, a voice recognition/analysis technology based on big data can elaborate customer services further, enhance customer satisfaction, and trigger constant interactions with customers. A customer center can be transformed to a hub of customer knowledge and the embodiment of business intelligence in the front line of business. This article is a case study on how the customer center of the K life insurance company regarding customer center operation collects and analyzes customer information and how it has established its voice recognition/analysis system based on big data to improve customer experience management. Factors affecting the successful introduction and implementation of voice recognition/analysis system to a firm, are examined.

• 동굴
• /
• 제88호
• /
• pp.1-7
• /
• 2008

Recently, it is one of very interest technology of Human Computer Interaction(HCI). Nowadays, it is easy to find out that, for example, inside SF movies people has talking to computer. However, there are difference between CPU language and ours. So, we focus on connecting to CPU. For 30 years many scientists experienced in that technology. But it is really difficult. Our project goal is making that CPU could understand human voice. First of all the signal through a voice sensor will move to BCD (binary code). That elevator helps out people who wants to move up and down. This product's point is related with people's safety. Using a PWM for motor control by ATmega16, we choose a DC motor to drive it because of making a regular speed elevator. Furthermore, using a voice identification module the elevator driven by voice sensor could operate well up and down perfectly from 1st to 10th floor by PWM control with ATmega16. And, it will be clearly useful for high-rise vertical shift with voice recognition sensor driven.

• 산업기술연구
• /
• 제33권A호
• /
• pp.55-60
• /
• 2013

In this paper, we proposed an efficient method for implementing a voice recognition elevator. Unlike the existing ones, the proposed system is based on the bluetooth communication and smartphones equipped with the google speech recognition software, which makes it possible that the speech recognition capability can be added to the previously installed elevators. In order to improve the recognition accuracy, instead of using the result of the google recognizer, we built a web server where the user data are accumulated and they are used for recognition error correction.

• 한국소프트웨어감정평가학회 논문지
• /
• 제15권2호
• /
• pp.73-83
• /
• 2019

음성인식은 최근 인공지능 기술과 접목되면서 오인식률이 크게 개선되었고 사용자 관점에서 효과적이고 효율적인 HMI(Human Machine Interface)를 제공하고 있다. 이러한 추세는 방위산업 분야에서도 반영되고 있고 특히 항공분야에서 F-35에 적용이 되었다. 하지만 이러한 기술에 대해서 품질평가를 위해 방위산업 특히 항공분야에는 객관적이고 평가 가능한 정량적인 모델이 필요하다. 본 연구에서는 이러한 음성인식의 소프트웨어 측면에서 항공기에 적용하기 위한 정량적인 평가 모델을 제시한다. 평가 모델 제시를 위해 음성인식 적용 기술과 ISO/IEC 25000(SQuaRE) 제품 품질 속성을 이용해 평가 항목을 추출한다. 이러한 두 가지 항목의 연계를 통해 정량적인 평가 모델을 제시하고 사례 연구를 활용해 평가 결과를 확인한다.

• 디지털융복합연구
• /
• 제11권1호
• /
• pp.225-236
• /
• 2013

음성인식 기술은 단순한 단어 인식을 넘어 자연스럽게 발성한 연속 음성도 인식할 수 있는 수준으로 발전해 왔다. 아이폰에 탑재된 자연어 음성인식 처리 소프트웨어인 시리(Siri)가 2010년에 발표되면서, 음성인식에 대한 연구가 관심을 받고 있다. 한국어 음성 인식 소프트웨어들은 대부분 단어 위주의 인식 서비스로 구성 되어 있으며, 잡음처리 및 음성 에너지 조절 기능들이 부족해 만족할 만한 인식률을 보이지 못하고 있다. 또한 요구된 발성 규칙을 따르지 못한 음성 질의들은 아예 처리하지 못하고 있는 실정이다. 본 논문에서는 이러한 현실적 어려움을 개선할 수 있도록 다목적 전처리 프로세서를 제안하였다. 이 처리기는 음성인식 엔진에 독립적이며, 잡음 제거 기능, 규칙에 따르지 않은 음성 질의도 처리 할 수 있는 핵심어 추출 기능, 그 핵심어를 수식하는 전술부 및 그 해당 음성 질의로부터 수행하기를 원하는 후술부 까지도 추출할 수 있는 기능을 갖추도록 하였다. 실험을 통해, 잡음 제거 효과 평가, 핵심어 인식 성공률, 연속음 인식 성공률을 측정하여 제안한 방법의 타당성을 확인하였다.

• 한국전자통신학회논문지
• /
• 제19권2호
• /
• pp.397-402
• /
• 2024

음성 인식과 모션 인식 기술은 다양한 스마트 디바이스에 적용되어 사용되고 있으나, 단순한 명령어 인식 형태로 구성되어 단순 기능으로 사용되고 있다. 인식 데이터에 대한 단순 기능에서 벗어나 다양한 분야에서 학습된 데이터를 기반으로 전문적인 명령어 수행 능력이 요구되고 있다. 현재 세계적으로 경쟁이 이루어지고 있는 생성형 AI를 활용하여 사용자에게 최적의 데이터를 제공하고, 음성 인식과 모션 인식을 통해 상호작용할 수 있는 시스템 플랫폼에 대한 연구가 진행되고 있다. 본 연구를 위해 설계한 주요 기술 프로세스는 음성 및 모션 인식 기능, AI 기술 적용, 양방향 커뮤니케이션 등 기술을 이용한 설계하였다. 본 논문에서는 AI 기술을 적용한 디바이스와 음성인식과 모션 인식 기술을 통해 디바이스와 사용자 간 양방향 커뮤니케이션을 다양한 입력방식에 의해 이루어질 수 있도록 하였다.

• 센서학회지
• /
• 제13권1호
• /
• pp.20-26
• /
• 2004

For the disabled persons who can't use their limbs, motorized wheelchair that is activated by voice recognition module employing speaker independent method, was implemented. The wireless voice transfer device was designed and employed for the user convenience. And the wheelchair was designed to operate using voice and keypad by selection of the user because they can manipulate it using keypad if necessary. The speaker independent method was used as the voice recognition module in order that anyone can manipulate the wheelchair in case of assistance. Using the implemented wheelchair, performance and motion of the system was examined and it has over than 97% of voice recognition rate and proper movements.

• 대한음성학회지:말소리
• /
• 제51호
• /
• pp.1-16
• /
• 2004

The aim of this paper is to analyze vowels in voice imitation and disguised voice, and to find the invariable phonetic features of the speaker. In this paper we examined the formants of monophthongs /a, u, i, o, {$\omega},{\;}{\varepsilon},{\;}{\Lambda}$/. The results of the present are as follows : $\circled1$ Speakers change their vocal tract features. $\circled2$ Vowels /a, ${\varepsilon}$, i/ appear to be proper for speaker recognition since they show invariable acoustic feature during voice modulation. $\circled3$ F1 does not change easily compared to higher formants. $\circled4$ F3-F2 appears to be constituent for a speaker identification in vowel /a/ and /$\varepsilon$/, and F4-F2 in vowel /i/. $\circled5$ Resulting of F-ratio, differences of each formants were more useful than individual formant of a vowel to speaker recognition.

• The Journal of the Acoustical Society of Korea
• /
• 제21권2E호
• /
• pp.98-104
• /
• 2002

The present study describes a combination method to recognize the human affective states such as anger, happiness, sadness, or surprise. For this, we extracted emotional features from voice signals and facial expressions, and then trained them to recognize emotional states using hidden Markov model (HMM) and neural network (NN). For voices, we used prosodic parameters such as pitch signals, energy, and their derivatives, which were then trained by HMM for recognition. For facial expressions, on the other hands, we used feature parameters extracted from thermal and visible images, and these feature parameters were then trained by NN for recognition. The recognition rates for the combined parameters obtained from voice and facial expressions showed better performance than any of two isolated sets of parameters. The simulation results were also compared with human questionnaire results.

• 한국산업융합학회 논문집
• /
• 제20권4호
• /
• pp.285-297
• /
• 2017

This study proposed a new approach to develop an Intelligent control algorithm for cooperative working of human and robot based on voice recognition. In general case of speaker verification, Gaussian Mixture Model is used to model the feature vectors of reference speech signals. On the other hand, Dynamic Time Warping based template matching techniques were presented for the voice recognition about several years ago. We converge these two different concepts in a single method and then implement in a real time voice recognition enough to make reference model to satisfy 95% of recognition performance. In this paper it was illustrated the reliability of voice recognition by simulation and experiments for humanoid robot with 18 joints.