• 제목/요약/키워드: One-Shot

검색결과 378건 처리시간 0.027초

Zero-shot voice conversion with HuBERT

  • Hyelee Chung;Hosung Nam
    • 말소리와 음성과학
    • /
    • 제15권3호
    • /
    • pp.69-74
    • /
    • 2023
  • This study introduces an innovative model for zero-shot voice conversion that utilizes the capabilities of HuBERT. Zero-shot voice conversion models can transform the speech of one speaker to mimic that of another, even when the model has not been exposed to the target speaker's voice during the training phase. Comprising five main components (HuBERT, feature encoder, flow, speaker encoder, and vocoder), the model offers remarkable performance across a range of scenarios. Notably, it excels in the challenging unseen-to-unseen voice-conversion tasks. The effectiveness of the model was assessed based on the mean opinion scores and similarity scores, reflecting high voice quality and similarity to the target speakers. This model demonstrates considerable promise for a range of real-world applications demanding high-quality voice conversion. This study sets a precedent in the exploration of HuBERT-based models for voice conversion, and presents new directions for future research in this domain. Despite its complexities, the robust performance of this model underscores the viability of HuBERT in advancing voice conversion technology, making it a significant contributor to the field.

퍼지 논리 알고리즘에 의한 사출제품의 미성형 해결 (Trouble Shooting of Short Shot in Injection Molding By Using Fuzzy Logic Algorithm)

  • Kang, Seong-Nam;Huh, Yong-Jeong;Cho, Hyun-Chan
    • 한국지능시스템학회:학술대회논문집
    • /
    • 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
    • /
    • pp.65-68
    • /
    • 2001
  • Short shot is a molded part that is incomplete since insufficient material was injected into the mold. Remedial actions to solve short shot can be done by injection molding experts based on their empirical knowledge. Modifying mold and part, changing resin to less viscous one, and adjusting process conditions are general remedies. Experts of injection molding might try to adjust process conditions such as mold temperature, melt temperature, injection time based on their empirical knowledge as the first remedy because adjustment of process conditions is the most economic way in time and cost. However it is difficult to find appropriate process conditions as they are highly coupled and there are so many elements to be considered. In this paper, a fuzzy logic algorithm has been proposed to find an appropriate mold temperature. With the percentage of the insufficient Quantity of an injection molded part, an appropriate mold temperature can be obtained by the fuzzy logic algorithm.

  • PDF

Effect of Shot Peening on Microstructural Evolution of 500-7 Ductile Cast Iron

  • Zhang, Yubing;Shin, Keesam
    • Applied Microscopy
    • /
    • 제48권3호
    • /
    • pp.73-80
    • /
    • 2018
  • Ductile cast iron is widely used for many automotive components due to its high wear resistance and fatigue resistance in addition to the low cost of fabrication. The improvement of wear resistance and fatigue properties is key to the life time extension and performance increase of the automobile parts. Surface nanocrystallization is a very efficient way of improving the performance of materials including the wear- and fatigue-resistance. Shot peening treatment, as one of the popular and economic surface modification methods, has been widely applied to various materials. In this study, ductile cast iron specimens were ultrasonic shot peening (USP) treated for 5 to 30 min using different ball size. The microstructures were then microscopically analyzed for determination of the microstructural evolution. After the USP treatment, the hardness of pearlite and ferrite increased, in which ball size is more effective than treatment time. With USP treatment, the graphite nodule count near the surface was decreased with grain refinement. The lager balls resulted in an increased deformation, whereas the smaller balls induced more homogenously refined grains in the deformation layer. In addition, formation of nanoparticles was formed in the surface layer upon USP.

A Study on Moldability by Using Fuzzy Logic Based Neural Network(FNN)

  • Kang, Seong Nam;Huh, Yong Jeong;Cho, Hyun Chan;Choi, Man Sung
    • 반도체디스플레이기술학회지
    • /
    • 제2권1호
    • /
    • pp.7-9
    • /
    • 2003
  • In order to predict the moldability of an injection molded part, a simulation of filling is needed. Short shot is one of the most frequent troubles encountered during injection molding process. The adjustment of process conditions is the most economic way to troubleshoot the problematic short shot in cost and time since the mold doesn't need to be modified at all. But it is difficult to adjust the process conditions appropriately in no times since it requires an empirical knowledge of injection molding. In this paper, the intelligent CAE system synergistically combines fuzzy-neural network(FNN) for heuristic knowledge with CAE programs for analytical knowledge. To evaluate the intelligent algorithms, a cellular phone flip has been chosen as a finite element model and filling analyses have been performed with a commercial CAE software. As the results, the intelligent CAE system drastically reduces the troubleshooting time of short shot in comparison with the expert's conventional way which is similar to the golden section search algorithm.

  • PDF

A Study on Moldability by Using Fuzzy Logic Based Neural Network(FNN)

  • Kang, Seong Nam;Huh, Yong Jeong;Choi, Man Sung
    • 한국반도체및디스플레이장비학회:학술대회논문집
    • /
    • 한국반도체및디스플레이장비학회 2002년도 추계학술대회 발표 논문집
    • /
    • pp.127-129
    • /
    • 2002
  • In order to predict the moldability of an injection molded part, a simulation of filling is needed. Short shot is one of the most frequent troubles encountered during injection molding process. The adjustment of process conditions is the most economic way to troubleshoot the problematic short shot in cost and time since the mold doesn't need to be modified at all. But it is difficult to adjust the process conditions appropriately in no times since it requires an empirical knowledge of injection molding. In this paper, the intelligent CAE system synergistically combines fuzzy-neural network(FNN) for heuristic knowledge with CAE programs for analytical knowledge. To evaluate the intelligent algorithms, a cellular phone flip has been chosen as a finite element model and filling analyses have been performed with a commercial CAE software. As the results, the intelligent CAE system drastically reduces the troubleshooting time of short shot in comparison with the expert's conventional way which is similar to the golden section search algorithm.

  • PDF

자동 촬영에 의한 파노라마 영상 생성 방법 (Panorama Image Construction Method By Automatic Shot)

  • 김태우
    • 한국산학기술학회논문지
    • /
    • 제8권6호
    • /
    • pp.1524-1529
    • /
    • 2007
  • 본 논문에서는 자동 촬영 파노라마 생성 방법을 제안한다. 기존에는 두 장의 파노라마 멤버들을 수동으로 촬영하여 파노라마 영상을 만드는 반면, 제안한 방법은 이동되는 카메라에서 파노라마 멤버들을 자동으로 촬영하여 파노라마 영상을 생성한다. 파노라마 멤버들은 카메라로부터 들어오는 영상 스트림에서 추적 영역을 자동으로 추적하여 촬영된다. 촬영된 멤버들은 추적 영역을 포함하는 정합 영역에 대해 불변 특징 방법을 적용한다. 이 방법은 파노라마 멤버들을 자동으로 촬영할 수 있고 파노라마 생성 속도가 빠른 장점이 있다. 실험에서 $320{\times}240$ 크기의 칼라 영상에 대해 제안한 방법의 처리 시간이 약 0.89초로 기존의 특징 기반 방법(6)에 비해 처리 속도가 약 2배 빠른 결과를 보였다.

  • PDF

일회성 시스템의 저장신뢰도 결정 모델에 관한 연구 (A Study on the Storage Reliability Determination Model for One-shot System)

  • 김동규;강운석;강성진
    • 한국경영과학회지
    • /
    • 제38권1호
    • /
    • pp.1-13
    • /
    • 2013
  • Some systems such as missiles and ammunitions are used only one time in combat or emergency situation. Predicting correct storage reliability is very important for those systems which are inspected periodically. Many researches have been done for predicting the storage reliability using generally exponential or Weibull failure distribution. However, recent studies show the hazard functions follow various types of failure distributions. So in this paper, we proposed a generalized model that measures the storage reliability regardless of type of failure distributions. And this model reflects inspection error and failures that might be occurred during periodical check and within storage term as well.

PFC용 부스트 컨버터의 병렬화에 의한 효율 개선 (An Improvement Parallel to the Efficiency of Boost Converter for Power Factor Correction)

  • 전내석;장수형;전일영;박영산;안병원;이성근;김윤식
    • 한국마린엔지니어링학회:학술대회논문집
    • /
    • 한국마린엔지니어링학회 2001년도 추계학술대회 논문집(Proceeding of the KOSME 2001 Autumn Annual Meeting)
    • /
    • pp.120-124
    • /
    • 2001
  • A new technique for improving the efficiency of single-phase high-frequency boost converter is proposed. This converter includes an additional low-frequency boost converter which is connected to the main high-frequency switching device in parallel. The additional converter is controlled at lower frequency. Most of the current flows in the low-frequency switch and so, high-frequency switching loss is greatly reduced accordingly Both switching device are controlled by a simple method; each controller consists of a one-shot multivibrator, a comparator and an AND gate. The converter works cooperatively in high efficiency and acts as if it were a conventional high-frequency boost converter with one switching device. The proposed method is verified by simulation. This paper describes the converter configuration and design, and discusses the steady-state performance concerning the switching loss reduction and efficiency improvement.

  • PDF