• 제목/요약/키워드: string mechanism

검색결과 48건 처리시간 0.021초

한국어 text-to-speech(TTS) 시스템을 위한 엔드투엔드 합성 방식 연구 (An end-to-end synthesis method for Korean text-to-speech systems)

  • 최연주;정영문;김영관;서영주;김회린
    • 말소리와 음성과학
    • /
    • 제10권1호
    • /
    • pp.39-48
    • /
    • 2018
  • A typical statistical parametric speech synthesis (text-to-speech, TTS) system consists of separate modules, such as a text analysis module, an acoustic modeling module, and a speech synthesis module. This causes two problems: 1) expert knowledge of each module is required, and 2) errors generated in each module accumulate passing through each module. An end-to-end TTS system could avoid such problems by synthesizing voice signals directly from an input string. In this study, we implemented an end-to-end Korean TTS system using Google's Tacotron, which is an end-to-end TTS system based on a sequence-to-sequence model with attention mechanism. We used 4392 utterances spoken by a Korean female speaker, an amount that corresponds to 37% of the dataset Google used for training Tacotron. Our system obtained mean opinion score (MOS) 2.98 and degradation mean opinion score (DMOS) 3.25. We will discuss the factors which affected training of the system. Experiments demonstrate that the post-processing network needs to be designed considering output language and input characters and that according to the amount of training data, the maximum value of n for n-grams modeled by the encoder should be small enough.

HRNCE 문법의 언어 생성력 (Language-generating-Power of HRNCE Grammars)

  • 정태의;박동선
    • 한국정보처리학회논문지
    • /
    • 제3권7호
    • /
    • pp.1659-1668
    • /
    • 1996
  • 스트링 문법은 그래프 언어와 부분집합인 스트링 언어를 생성하는 반면 그래프 문법은 그래프 언어를 생성한다. 그래프 문법 모델중 가장 성공적인 것중의 하나인 NLC문법은 노드 레이블을 이용하여 하나의 노드를 하나의 그래프로 치환 하므로서 그래프를 생성한다. 그래프를 포함하는 하아피그래프를 생성하는 하이퍼그래프 문법 모델로는 미리 정해놓은 접착점들의 순서를 이용하여 하나의 하이퍼에지를 하나의 하이퍼그래프로 치환하는 CFHG 문법, CFHG 문법의 확장형으로서 복제 및 삭제의 방벙 을 통해 하이페이지와 그것이 포함하고 있는 노드들, 즉 하나의 핸들을 하나의 하아 퍼그래프로 차환하는 HH 문법, 그리고 eNCE방식을 이용하여 하나의 핸들을 하나의 하아퍼그래프로 치환하는 HRNCE 문법 등이 소개되었다. 본 논문에서는 HRNCE 문법이 생성하는 그래프 언어와 위에서 언급된 기타의 그래프 문법들이 생성하는 그래프 언어 들을 비교하므로서 HRNCE 문법과 타 그래프 문법들의 언어 생성력을 비교 검토하고자 한다.

  • PDF

통합 방식을 이용한 대보표 악보의 XML 스키마 모델 (XML Schema Model of Great Staff Music Score using the Integration Method)

  • 김정희;곽호영
    • 한국정보통신학회논문지
    • /
    • 제7권2호
    • /
    • pp.302-313
    • /
    • 2003
  • 현재 악보에 대한 DTD(Document Type Definition) 정의는 응용에 따라 그 종류가 다양하게 연구되어 왔으며 또한 정의된 DTD를 XML Schema로 자동 변환하는 방법들이 진행되고 있다. 그리고 연구되어온 DTD 정의 구조는 악보 상에 존재하는 음악 정보들이 각각 독려된 형식을 유지하도록 표현하는 데 집중되고 있다. 따라서 본 논문에서는 악보의 기본 구성 요소가 소절(Measure:마디)인 점을 이용하여 소절 단위내 관련된 정보를 일련의 연속 문자열 값으로 표현하는 방법을 제안하고 이를 위한 XML Schema를 모델링 하였다 또한 제안된 방법에 의해 표현된 XML 인스턴스에서 음악 정보를 추출하는 메카니즘을 제시하였다. 그 결과 연속된 문자열 값들을 가지는 XML Schema 정의가 가능했으며, 제안된 방식을 적용한 인스턴스는 기존 방식의 인스턴스들 보다 검색식(XPATH) 표현의 단순화 및 검색 단계를 줄여 그 효율성이 기대된다. 또한 사람에 의한 직접 표현 능력이 가능하며, 인스턴스 크기가 작아짐을 알 수 있었다.

방풍(防風)과 해방풍(海防風) 중 뇌경색 연구에 더욱 적합한 약재 선정을 위한 네트워크 약리학적 분석 (Network pharmacoligical analysis for selection between Saposhnikoviae Radix and Glehniae Radix focusing on ischemic stroke)

  • 진예진;임세현;조수인
    • 대한한의학방제학회지
    • /
    • 제31권3호
    • /
    • pp.171-182
    • /
    • 2023
  • Objectives : Saposhnikoviae Radix (SR) and Glehniae Radix (GR) have been frequently used in traditional medicine to treat diseases related to 'wind' syndrome, but there have been cases where it has been mixed in a state where the plant of origin is not clear. In this study, to select materials for conducting preclinical cerebral infarction research, the network pharmacology analysis method was used to select suitable medicinal materials for the study. Methods : In this study, a Traditional Chinese Medicine Systems Pharmacology Database and Analysis Platform (TCMSP) based network pharmacology analysis method was used, and oral bioavailability (OB), drug likeness (DL), Caco-2 and BBB permeability were utilized to select compounds with potential activity. For the values of each variable used in this study, OB ≥ 20%, DL ≥ 0.18, Caco-2 ≥ 0, and BBB ≥ -0.3 were applied, then networks of bioactive compounds, target proteins, and target diseases was constructed. STRING database was used to construct a protein-protein interaction network. Results : It was confirmed that SR rather than GR has various target proteins and target diseases based on network pharmacological analysis using TCMSP database. And it was analyzed that the bioactive compounds only in SR act more on neurovascular diseases, and both drugs are expected to be effectively used for cardiovascular diseases. Conclusions : In our future study, SR will be used in an ischemic stroke mouse model, and the mechanism of action will be explored focusing on apoptosis and cell proliferation.

패턴 인식을 위한 진화 셀룰라 분류기 (Evolvable Cellular Classifiers for pattern Recognition)

  • 주재호;신윤철;강훈
    • 한국지능시스템학회논문지
    • /
    • 제10권4호
    • /
    • pp.379-389
    • /
    • 2000
  • A cellular automaton is well-known for self-organizing and dynamic behavions in the filed of artifial life. This paper addresses a new neuronic architecture called an evolvable celluar classifier which evolves with the genetic rules (chromosomes) in the non-uniform cellular automata. An evolvable cellular classifier is primarily based on cellular programming, but its mechanism is simpler becaise it utilizes only mutations for the main genetic operators and resmbles the Hopfield network. Therefore, the desirable bit-patterns could be obtained through evolutionary processes for just one individual agent, As a rusult, an evolvable hardware is derived which is applicable to clessification of bit-string information.

  • PDF

패턴 인식을 위한 진화 셀룰라 분류기 (Evolvable Cellular Classifiers for Pattern Recognition)

  • Ju, Jae-ho;Shin, Yoon-cheol;Hoon Kang
    • 한국지능시스템학회:학술대회논문집
    • /
    • 한국퍼지및지능시스템학회 2000년도 춘계학술대회 학술발표 논문집
    • /
    • pp.236-240
    • /
    • 2000
  • A cellular automaton is well-known for self-organizing and dynamic behaviors in the field of artificial life. This paper addresses a new neuronic architecture called an evolvable cellular classifier which evolves with the genetic rules (chromosomes) in the non-uniform cellular automata. An evolvable cellular classifier is primarily based on cellular programing, but its mechanism is simpler because it utilizes only mutations for the main genetic operators and resembles the Hopfield network. Therefore, the desirable hi t-patterns could be obtained through evolutionary processes for just one individual agent. As a result, an evolvable hardware is derived which is applicable to classification of bit-string information.

  • PDF

Double-String Model for Auditory Transduction of Drosophila

  • Lee, Woo Seok;Ahn, Kang-Hun;Lee, Jeongmi;Chung, Yun Doo;Mhatre, Natasha;Robert, Daniel
    • Journal of the Korean Physical Society
    • /
    • 제73권9호
    • /
    • pp.1225-1229
    • /
    • 2018
  • The Drosophila auditory system consists of four large basal segments: the arista, the funiculus, the pedicel, and the scape. When an acoustic stimulus is applied to the arista and the funiculus their mechanical vibrations are transmitted to chordotonal neurons in Johnston's organ where mechanoelectric transduction arises. We study the mechanotransduction mechanism in the Drosophila auditory system by using a laser Doppler vibrometer (LDV) and extracellular electrophysiology. We find that large and small peaks appear alternatively and that the antenna vibration is asymmetric depending on whether the pedicel and the scape are fixed. Interestingly, we find that this asymmetric vibration accompanies the alternating neural peak structure. Here, we propose a mathematical model to explain the alternating peak structure by using a model consisting of two opposing neurons that are modeled as strings. Generally, strings have tension only when they are elongated. This property allows the alternating neural peaks for asymmetric antenna motion.

룰 기반 웹 IDS 시스템을 위한 효율적인 웹 로그 전처리 기법 설계 및 구현 (Design and Implementation of Advanced Web Log Preprocess Algorithm for Rule based Web IDS)

  • 이형우
    • 인터넷정보학회논문지
    • /
    • 제9권5호
    • /
    • pp.23-34
    • /
    • 2008
  • 웹 기반 서비스가 다양한 형태로 제공되면서 웹 서비스 사용자 수는 꾸준히 증가하고 있다. 그러나 웹 서버에 대한 SQL Injection, Parameter Injection 및 DoS 등의 공격 등의 취약점이 발견되고 있다. 이와 같은 형태의 웹 공격에 능동적으로 대응하기 위해 현재 웹 IDS 시스템을 구축하여 룰 기반 대응 시스템을 구축하고 있으나, 웹 서버에서 생성되는 로그 정보에 대한 전처리 과정 없이 룰 기반 IDS 시스템이 구동되기 때문에 효율적인 웹 공격 대응체계가 구축되지 못하고 있다. 이에 본 연구에서는 웹 로그 정보를 웹 IDS 기반 공격 탐지 시스템의 룰 비교 특성에 적합한 형태로 전처리하는 알고리즘을 제시하고 이를 구현하였다. 제안한 알고리즘은 웹 로그 정보에 대한 필드 단위 파싱 및 중복 문자열 처리 과정을 고속으로 수행하여 대용량의 로그 처리시 성능을 향상시켜 개선된 웹 IDS 시스템 구축이 가능하다.

  • PDF

Integrative applications of network pharmacology and molecular docking: An herbal formula ameliorates H9c2 cells injury through pyroptosis

  • Zhongwen Qi;Zhipeng Yan;Yueyao Wang;Nan Ji;Xiaoya Yang;Ao Zhang;Meng Li;Fengqin Xu;Junping Zhang
    • Journal of Ginseng Research
    • /
    • 제47권2호
    • /
    • pp.228-236
    • /
    • 2023
  • Background: QiShen YiQi pills (QSYQ) is a Traditional Chinese Medicine (TCM) formula, which has a significant effect on the treatment of patients with myocardial infarction (MI) in clinical practice. However, the molecular mechanism of QSYQ regulation pyroptosis after MI is still not fully known. Hence, this study was designed to reveal the mechanism of the active ingredient in QSYQ. Methods: Integrated approach of network pharmacology and molecular docking, were conducted to screen active components and corresponding common target genes of QSYQ in intervening pyroptosis after MI. Subsequently, STRING and Cytoscape were applied to construct a PPI network, and obtain candidate active compounds. Molecular docking was performed to verify the binding ability of candidate components to pyroptosis proteins and oxygen-glucose deprivation (OGD) induced cardiomyocytes injuries were applied to explore the protective effect and mechanism of the candidate drug. Results: Two drug-likeness compounds were preliminarily selected, and the binding capacity between Ginsenoside Rh2 (Rh2) and key target High Mobility Group Box 1 (HMGB1)was validated in the form of hydrogen bonding. 2 μM Rh2 prevented OGD-induced H9c2 death and reduced IL-18 and IL-1β levels, possibly by decreasing the activation of the NLRP3 inflammasome, inhibiting the expression of p12-caspase1, and attenuating the level of pyroptosis executive protein GSDMD-N. Conclusions: We propose that Rh2 of QSYQ can protect myocardial cells partially by ameliorating pyroptosis, which seems to have a new insight regarding the therapeutic potential for MI.

유전알고리즘을 이용한 편측식 선형유도전동기의 최적설계 (Optimal Design of Single-sided Linear Induction Motor Using Genetic Algorithm)

  • 류근배;최영준;김창업;김성우;임달호
    • 대한전기학회:학술대회논문집
    • /
    • 대한전기학회 1993년도 하계학술대회 논문집 B
    • /
    • pp.923-928
    • /
    • 1993
  • Genetic algorithms are powerful optimization methods based on the mechanism of natural genetics and natural selection. Genetic algorithms reduce chance of searching local optima unlike most conventional search algorithms and especially show good performances in complex nonlinear optimization problems because they do not require any information except objective function value. This paper presents a new model based on sexual reproduction in nature. In the proposed Sexual Reproduction model(SR model), individuals consist of the diploid of chromosomes, which are artificially coded as binary string in computer program. The meiosis is modeled to produce the sexual cell(gamete). In the artificial meiosis, crossover between homologous chromosomes plays an essential role for exchanging genetic informations. We apply proposed SR model to optimization of the design parameters of Single-sided Linear Induction Motor(SLIM). Sequential Unconstrained Minimization Technique(SUMT) is used to transform the nonlinear optimization problem with many constraints of SLIM to a simple unconstrained problem, We perform optimal design of SLIM available to FA conveyer systems and discuss its results.

  • PDF