• Annual Conference on Human and Language Technology
• /
• 2006.10e
• /
• pp.183-188
• /
• 2006

본 논문에서 위치 정보 데이터란 텔레메틱스 분야의 응용을 위하여 웹상에서 수집한 Point-of-Interest (POI) 데이터로서 행정구역 및 지명 인명, 상호명과 같은 위치 검색에 사용되는 어휘로 구성된다. 본 논문은 음성 인식 시스템을 구성하는 발음 사전의 개발에 관한 것으로 250k 위치 정보데이터로부터 운율어를 이용하여 불규칙 발음과 발음 변이를 포함하는 가능한 모든 발음을 생성하는 방법을 제안하는 것을 목적으로 한다. 원래 모든 POI 는 한 번씩만 데이터에 포함되어 있으므로, 그 가운데 불규칙 발음을 포함하는 POI를 검출하거나 발음을 생성하기 위해서는 각각의 POI 하나하나를 일일이 검토하는 방법밖에 없는데, 대부분의 POI 가 복합명사구로 이루어졌다는 점에 착안하여 운율어를 이용한 결과, 불규칙 발음 검출과 다중 발음 생성을 효율적으로 수행할 수 있었다. 이러한 연구는 음성처리 영역에서는 위치정보데이터의 음성인식 성능을 향상하는 데 직접적인 기여를 할 수 있고, 무엇보다도 음성학과 음운론 이론을 음성 인식 분야에 접목한 학제적 연구로서 그 의미가 있다고 할 수 있다.

• 한국정보컨버전스학회:학술대회논문집
• /
• 2008.06a
• /
• pp.53-56
• /
• 2008

This paper reports on the ACQA(Animated agent for Conversational Question Answering) project conducted at LIMSI. The aim is to design an expressive animated conversational agent(ACA) for conducting research along two main lines: 1/ perceptual experiments(eg perception of expressivity and 3D movements in both audio and visual channels): 2/ design of human-computer interfaces requiring head models at different resolutions and the integration of the talking head in virtual scenes. The target application of this expressive ACA is a real-time question and answer speech based system developed at LIMSI(RITEL). The architecture of the system is based on distributed modules exchanging messages through a network protocol. The main components of the system are: RITEL a question and answer system searching raw text, which is able to produce a text(the answer) and attitudinal information; this attitudinal information is then processed for delivering expressive tags; the text is converted into phoneme, viseme, and prosodic descriptions. Audio speech is generated by the LIMSI selection-concatenation text-to-speech engine. Visual speech is using MPEG4 keypoint-based animation, and is rendered in real-time by Virtual Choreographer (VirChor), a GPU-based 3D engine. Finally, visual and audio speech is played in a 3D audio and visual scene. The project also puts a lot of effort for realistic visual and audio 3D rendering. A new model of phoneme-dependant human radiation patterns is included in the speech synthesis system, so that the ACA can move in the virtual scene with realistic 3D visual and audio rendering.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.3
• /
• pp.63-69
• /
• 2001

A corpus-based lip sync algorithm for synthesizing natural face animation is proposed in this paper. To get the lip parameters, some marks were attached some marks to the speaker's face, and the marks' positions were extracted with some Image processing methods. Also, the spoken utterances were labeled with HTK and prosodic information (duration, pitch and intensity) were analyzed. An audio-visual corpus was constructed by combining the speech and image information. The basic unit used in our approach is syllable unit. Based on this Audio-visual corpus, lip information represented by mark's positions was synthesized. That is. the best syllable units are selected from the audio-visual corpus and each visual information of selected syllable units are concatenated. There are two processes to obtain the best units. One is to select the N-best candidates for each syllable. The other is to select the best smooth unit sequences, which is done by Viterbi decoding algorithm. For these process, the two distance proposed between syllable units. They are a phonetic environment distance measure and a prosody distance measure. Computer simulation results showed that our proposed algorithm had good performances. Especially, it was shown that pitch and intensity information is also important as like duration information in lip sync.

• Journal of the Korean Society for Library and Information Science
• /
• v.46 no.1
• /
• pp.163-181
• /
• 2012

We investigated how useful video tags were in summarizing video speech and how valuable positional information was for speech summarization. Furthermore, we examined the similarity among sentences selected for a speech summary to reduce its redundancy. Based on such analysis results, we then designed and evaluated a method for automatically summarizing speech transcripts using a modified Maximum Marginal Relevance model. This model did not only reduce redundancy but it also enabled the use of social tags, title words, and sentence positional information. Finally, we compared the proposed method to the Extractor system in which key sentences of a video speech were chosen using the frequency and location information of speech content words. Results showed that the precision and recall rates of the proposed method were higher than those of the Extractor system, although there was no significant difference in the recall rates.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.6
• /
• pp.277-281
• /
• 2006

This paper presents a method for the automatic recognition of pitch accents over syllables. The method that we propose is based on the time-delay recursive neural network (TDRNN). which is a neural network classifier with two different representation of dynamic context: the delayed input nodes allow the representation of an explicit trajectory F0(t) along time. while the recursive nodes provide long-term context information that reflects the characteristics of pitch accentuation in spoken English. We apply the TDRNN to pitch accent recognition in two forms: in the normal TDRNN. all of the prosodic features (pitch. energy, duration) are used as an entire set in a single TDRNN. while in the distributed TDRNN. the network consists of several TDRNNs each taking a single prosodic feature as the input. The final output of the distributed TDRNN is weighted sum of the output of individual TDRNN. We used the Boston Radio News Corpus (BRNC) for the experiments on the speaker-independent pitch accent recognition. π 1e experimental results show that the distributed TDRNN exhibits an average recognition accuracy of 83.64% over both pitch events and non-events.

• The KIPS Transactions:PartB
• /
• v.15B no.6
• /
• pp.595-602
• /
• 2008

The focused word of each sentence is a help in recognizing and understanding spoken Korean. To find the method of focused word spotting at spoken speech signal, we made an analysis of the average and variance of Fundamental Frequency and the average energy extracted from a focused word and the other words in a sentence by experiments with the speech data from 100 spoken sentences. The result showed that focused words have either higher relative average F0 or higher relative variances of F0 than other words. Our findings are to make a contribution to getting prosodic characteristics of spoken Korean and keyword extraction based on natural language processing.

• Journal of the Korean Society for information Management
• /
• v.29 no.4
• /
• pp.191-208
• /
• 2012

Two fundamental aspects of speech summary generation are the extraction of key speech content and the style of presentation of the extracted speech synopses. We first investigated whether acoustic features (speaking rate, pitch pattern, and intensity) are equally important and, if not, which one can be effectively modeled to compute the significance of segments for lecture summarization. As a result, we found that the intensity (that is, difference between max DB and min DB) is the most efficient factor for speech summarization. We evaluated the intensity-based method of using the difference between max-DB and min-DB by comparing it to the keyword-based method in terms of which method produces better speech summaries and of how similar weight values assigned to segments by two methods are. Then, we investigated the way to present speech summaries to the viewers. As such, for speech summarization, we suggested how to extract key segments from a speech video efficiently using acoustic features and then present the extracted segments to the viewers.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.337-340
• /
• 2004

The exactly reproduced prosody of a TTS system is one of the key factors that affect the naturalness of synthesized speech. In general, rules about prosody had been gathered either from linguistic knowledge or by analyzing the prosodic information from natural speech. But these could not be perfect and some of them could be incorrect. So we proposed artificial neural network(ANN)s that can be trained to team the prosody of natural speech and generate it. In learning phase, let ANNs learn the pitch and energy contour of center phoneme by applying a string of phonemes in a sentence to ANNs and comparing the output pattern with target pattern and making adjustment in weighting values to get the least mean square error between them. In test phase, the estimation rates were computed. We saw that ANNs could generate the prosody of a sentence.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.1
• /
• pp.96-107
• /
• 1996

A voice personality transformation algorithm using orthogonal vector space conversion is proposed in this paper. Voice personality transformation is the process of changing one person's acoustic features (source) to those of another person (target). In this paper, personality transformation is achieved by changing the LPC cepstrum coefficients, excitation spectrum and pitch contour. An orthogonal vector space conversion technique is proposed to transform the LPC cepstrum coefficients. The LPC cepstrum transformation is implemented by principle component decomposition by applying the Karhunen-Loeve transformation and minimum mean-square error coordinate transformation(MSECT). Additionally, we propose a pitch contour modification method to transform the prosodic characteristics of any speaker. To do this, reference pitch patterns for source and target speaker are firstly built up, and speaker's one. The experimental results show the effectiveness of the proposed algorithm in both subjective and objective evaluations.

• Journal of Digital Contents Society
• /
• v.16 no.2
• /
• pp.291-298
• /
• 2015

In a fast-paced information technology environment, consumption of video content is changing from one-way television viewing to VOD (Video on Demand) playing anywhere, anytime, on any device. This video-watching trend gives additional importance to videos with fine-speed-control, in addition to the strength of the digital video signal. Currently, many video players provide a fine-speed-control function which can speed up the video to skip a boring part, or slow it down to focus on an exciting scene. The audio information is just as important as the visual information for understanding the content of the speed-controlled video. Thus, a number of algorithms for fine-speed-control video-playing technologies have been proposed to solve the pitch distortion in the audio-processing area. In this study, well-known techniques for prosodic modification of speech signals, WSOLA (Waveform-Similarity-Based Overlap-Add), have been applied to analyze users' needs for fine-speed-control video playing. By surveying the users' preferred speeds on categorized video content and analyzing the results, this paper proposes that various fine-speed adjustments are needed to accommodate users' preferred video consumption.