http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
분산음성인식 환경에서 서버에서의 스케일러블 고품질 음성복원
윤재삼(Jae Sam Yoon),김홍국(Hong Kook Kim),강병옥(Byung-Ok Kang) 대한전자공학회 2007 대한전자공학회 학술대회 Vol.2007 No.7
In this paper, we propose a scalable high-quality speech reconstruction method for distributed speech recognition (DSR). It is difficult to reconstruct speech of high quality with MFCCs at the DSR server. Depending on the bit-rate available by the DSR system, we can send additional information associated with speech coding to the DSR server, where the bit-rate is variable from 4.8 kbit/s to 11.4 kbit/s. The experimental results show that the speech quality reproduced by the proposed method when the bit-rate is 11.4 kbit/s is comparable with that of ITU-T G.729 under both ideal channel and frame error channel conditions while the performance of DSR is maintained to that of wireline speech recognition.
이종삼,윤재만,전중기,고성경 대구대학교 인문과학연구소 2007 人文科學硏究 Vol.30 No.-
We investigated that the effect of passive smoking on percent oxygen saturation (%SaO2) and blood pressure (BP). Total 16 healthy male subjects were recruited, and subjects were allocated into one of two groups: either smoker group (n=8) or non-smoker group (n=8). All subjects were stayed in sealed laboratory room until all the experiment completed. Each subject in smoker group have smoked every one hour, and %SaO2 in arterial blood, BP, heart rate (HR), lung vital capacity (VC), and blood glucose and lactate were measured. BP was increased more rapidly in non-smoker group than smoker group during experimental period. SaO2 was significantly lower in non-smoker group compared to smoker group. There was no significant change in HR, and blood glucose and lactate. It was not also significantly changed in VC. In conclusion, passive smoking can deteriorate in %SaO2 and BP in nonsmoker healthy people, and smoking should be prohibited in any public area as well as room in which multi-persons work together.
네트워크 환경에서 서버용 음성 인식을 위한 MFCC 기반 음성 부호화기 설계
이길호,윤재삼,오유리,김홍국,Lee, Gil-Ho,Yoon, Jae-Sam,Oh, Yoo-Rhee,Kim, Hong-Kook 대한음성학회 2005 말소리 Vol.54 No.-
Existing standard speech coders can provide speech communication of high quality while they degrade the performance of speech recognition systems that use the reconstructed speech by the coders. The main cause of the degradation is that the spectral envelope parameters in speech coding are optimized to speech quality rather than to the performance of speech recognition. For example, mel-frequency cepstral coefficient (MFCC) is generally known to provide better speech recognition performance than linear prediction coefficient (LPC) that is a typical parameter set in speech coding. In this paper, we propose a speech coder using MFCC instead of LPC to improve the performance of a server-based speech recognition system in network environments. However, the main drawback of using MFCC is to develop the efficient MFCC quantization with a low-bit rate. First, we explore the interframe correlation of MFCCs, which results in the predictive quantization of MFCC. Second, a safety-net scheme is proposed to make the MFCC-based speech coder robust to channel error. As a result, we propose a 8.7 kbps MFCC-based CELP coder. It is shown from a PESQ test that the proposed speech coder has a comparable speech quality to 8 kbps G.729 while it is shown that the performance of speech recognition using the proposed speech coder is better than that using G.729.
오유리 (Yoo Rhee Oh),윤재삼 (Jae Sam Yoon),박지훈 (Ji Hun Park),김민아 (Mina Kim),김홍국 (Hong Kook Kim),공동건 (Donggeon Kong),명현 (Hyun Myung),방석원(Seokwon Bang) 한국HCI학회 2008 한국HCI학회 학술대회 Vol.2008 No.2
본 논문에서는 Call-and-Come 서비스를 제공하는 가정용 로봇의 호출음 등록 및 인식 시스템을 구축하고, 음성 기반의 효율적인 로봇 호출음 등록 및 인식 알고리즘을 제안한다. 본 논문에서는 음성을 이용하여 로봇 호출음을 효율적으로 등록하기 위해 monophone 음향모델을 이용하여 탐색 범위를 줄이고, 줄어든 탐색 범위 내에서 triphone 음향모델을 이용하여 호출음을 등록을 한다. 또한, 잘못된 호출이 인식되는 것을 줄이기 위한 발화 검증에 필요한 파라미터를 구한다. 원거리 음성인식률을 향상시키기 위해서 근거리 음성에 최적화된 음향모델을 원거리 음성 데이터베이스로 적응시켰으며, 마이크로폰 배열을 이용하여 사용자의 위치를 추정한다. 제안한 시스템의 성능 측정을 위해 수행된 로봇 호출음에 대한 등록 및 인식 실험에서 98.3%의 음성 인식률을 얻었다. We propose an efficient robot name registration and recognition method in order to enable a Call-and-Come service for home robots. In the proposed method for the name registration, the search space is first restricted by using monophone-based acoustic models. Second, the registration of robot names is completed by using triphone-based acoustic models in the restricted search space. Next, the parameter for the utterance verification is calculated to reduce the acceptance rate of false calls. In addition, acoustic models are adapted by using a distance speech database to improve the performance of distance speech recognition, Moreover, the location of a user is estimated by using a microphone array. The experimental result on the registration and recognition of robot names shows that the word accuracy of speech recognition is 98.3%.
거주 형태에 따른 에너지 섭취량과 소비량의 균형도 조사
박순목,고성경,남인수,윤재만,임승현,전중기,이종삼 대구대학교 인문과학연구소 2008 人文科學硏究 Vol.31 No.-
We investigated that the effects of residential type on energy balance in college students. Total sixteen college students were participated in this study, all subjects were assigned one of three groups: either school attendee students group, self-governed living students group, dormitory students group. Routine physical activity level (for 5 days including three weekday and two weekend) and food intake were surveyed. For investigation of degree of physical activity, all subjects were requested to record on their physical movements as possible as detail should be obtained. To all subjects, five-day dietary log form was given, and used for examining of calorie intake from their routine diet. There was no statistical difference in energy intake and consumption in each. However energy consumption was significantly higher than energy uptake in school attendee students group. All other groups were shown a similar energy values between energy uptake and consumption. There were no significant differences in energy intake and consumption in any of experimental groups when comparisons were made between weekdays and weekend. As far as energy uptake was concerned it was no statistical difference in any of major nutrients among groups. In conclusion, partial imbalance was found between energy intake and uptake in school attendee groups. This may be due to their more active life style than other groups'. In future studies, better controlled study should be performed not only more subjects are recruited but also minor nutrients are included for examining of energy balance.
오유리(Yoo Rhee Oh),윤재삼(Jae Sam Yoon),박지훈,김민아(Mina Kim),김홍국(Hong Kook Kim) 대한전자공학회 2007 대한전자공학회 학술대회 Vol.2007 No.7
In this paper, we implement an automatic distance speech recognition system for voiced-enabled services. We first construct a baseline automatic speech recognition (ASR) system, where acoustic models are trained from speech utterances spoken by using a cross-talking microphone, In order to improve the performance of the baseline ASH using distance speech, the acoustic models are adapted to adjust the spectral characteristics of speech according to different microphones and the environmental mismatches between cross-talking and distance speech. Next, we develop a voice activity detection algorithm for distance speech, We compare the performance of the baseline system and the developed ASH system on a task of PBW (Phonetically Balanced Word) 452. As a result, it is shown that the developed ASH system provides the average word error rate (WER) reduction of 30.6 % compared to the baseline ASH system.