
http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
코퍼스 기반 음성합성기를 위한 합성단위 경계 스펙트럼 평탄화 알고리즘
김상진,장경애,한민수,Kim Sang-Jin,Jang Kyung Ae,Hahn Minsoo 대한음성학회 2005 말소리 Vol.56 No.-
Speech unit concatenation with a large database is presently the most popular method for speech synthesis. In this approach, the mismatches at the unit boundaries are unavoidable and become one of the reasons for quality degradation. This paper proposes an algorithm to reduce undesired discontinuities between the subsequent units. Optimal matching points are calculated in two steps. Firstly, the fullback-Leibler distance measurement is utilized for the spectral matching, then the unit sliding and the overlap windowing are used for the waveform matching. The proposed algorithm is implemented for the corpus-based unit concatenating Korean text-to-speech system that has an automatically labeled database. Experimental results show that our algorithm is fairly better than the raw concatenation or the overlap smoothing method.
연결형 합성음성을 이용한 경보음의 주관적 위급도 정량화
장필식,이경태,Jang, Pil-Sik,Lee, Gyeong-Tae 대한인간공학회 2006 대한인간공학회지 Vol.25 No.2
This paper presents an experimental study of the factors modulating the urgency perception of voice alarm generated by concatenative synthesizers. Four experiments were conducted using psycho-physical approach in which 105 participants made magnitude estimation for urgency perception of various voice alarm stimuli. Experiment 1 identified 6 acoustic and non-acoustic factors modulating the perceived urgency of synthesized voice alarm. Experiment 2, 3 and 4 quantified the relations between the objective changes in each of the quantifiable parameters and the subjective changes in urgency perception. This research has implications for the design and implementation of synthesized voice alarm systems where urgency mapping is required.
Auspex: Ecosystemic Emergence as Generative Soundscape
Mike Cassidy(마이크 캐시디),Kristian North(크리스찬 노스) 한국전자음악협회 2023 에밀레 Vol.21 No.-
This paper presents an overview of Auspex, an agent-based artificial life ecosystem generating multichannel soundscapes from corpus-based concatenative synthesis techniques aligned with the acoustic niche hypothesis. Insights into its development are derived from the review of two prototypical systems; BoidGran, a boid-driven granular synthesizer, and Swarmscape, an audio-visual performance system that emphasizes the acoustic emergence of swarming bodies. Analysis includes conceptual approaches in the contextualization of the system’s output.