• 비교문화연구
• /
• 제24권
• /
• pp.157-169
• /
• 2011

Compared to studies on prosodic effects on the comprehension of syntactic ambiguity in English, there are relatively few that investigated prosodic effects in East-Asian languages. This study examined the role of prosodic information in processing syntactically ambiguous sentences in Japanese. For syntactically ambiguous sentences containing relative clauses, this paper investigated whether prosodic information is immediately available during the process of these ambiguous sentences. Results from an auditory comprehension experiment with an on-line, cross-modal naming task seemingly suggest that contrary to the findings from the off-line study that examined the same constructions, prosodic information may not be immediately available to Japanese listeners. A possible account for failure to obtain effects of prosodic information is provided.

• 말소리와 음성과학
• /
• 제3권4호
• /
• pp.71-81
• /
• 2011

This paper investigated how prosodic position and word type affect the phonetic structure of Korean coronal stops. Initial segments of prosodic domains were known to be more strongly articulated and longer relative to prosodic domain-medial segments. However, there are few studies examining whether the properties of prosodic domain-initial segments are affected by the information content of words (real vs. nonsense words). In addition, since the scope of domain-initial effect was known to be local to the initial consonant and the effects on the following vowel have been found to be limited, it is thus worth examining whether the prosodic domain-initial effect extends into the vowel after the initial consonant in a systematic way across different prosodic domains. The acoustic properties of Korean coronal stops (lenis /t/, aspirated /$t^h$/, and tense /t'/) were compared across Intonational Phrase, Phonological Phrase and Word-initial positions both in real and nonsense words. The durational intervals such as VOT and CV duration were cumulatively lengthened for /t/ and /$t^h$/ in the higher prosodic domain-initial positions. However, tense stop /t'/ did not show any variation as a function of prosodic position and word type. The domain-initial lenis stop showed significantly longer duration in nonsense words than in real words. But the prosodic domain-initial effect was not found in the properties of F0 and [H1-H2] of the vowel after initial stops. The present study provided evidence that speakers tend to enhance speech clarity when there is less contextual information as in prosodic domain-initial position and in nonsense words.

• 대한음성학회지:말소리
• /
• 제66호
• /
• pp.117-130
• /
• 2008

This paper presents a method for automatically detecting Korean prosodic boundaries using both acoustic and grammatical information for the performance improvement of speech information processing systems. While most of previous works are solely based on grammatical information, our method utilizes not only grammatical information constructed by a Maximum-Entropy-based grammar model using 10 grammatical features, but also acoustical information constructed by a GMM-based acoustic model using 14 acoustic features. Given that Korean prosodic structure has two intonationally defined prosodic units, intonation phrase (IP) and accentual phrase (AP), experimental results show that the detection rate of AP boundaries is 82.6%, which is higher than the labeler agreement rate in hand transcribing, and that the detection rate of IP boundaries is 88.7%, which is slightly lower than the labeler agreement rate.

• The Journal of the Acoustical Society of Korea
• /
• 제28권2E호
• /
• pp.43-50
• /
• 2009

To get more natural synthetic speech generated by a Korean TTS (Text-To-Speech) system, we have to know all the possible prosodic rules in Korean spoken language. We should find out these rules from linguistic, phonetic information or from real speech. In general, all of these rules should be integrated into a prosody-generation algorithm in a TTS system. But this algorithm cannot cover up all the possible prosodic rules in a language and it is not perfect, so the naturalness of synthesized speech cannot be as good as we expect. ANNs (Artificial Neural Networks) can be trained to learn the prosodic rules in Korean spoken language. To train and test ANNs, we need to prepare the prosodic patterns of all the phonemic segments in a prosodic corpus. A prosodic corpus will include meaningful sentences to represent all the possible prosodic rules. Sentences in the corpus were made by picking up a series of words from the list of PB (phonetically Balanced) isolated words. These sentences in the corpus were read by speakers, recorded, and collected as a speech database. By analyzing recorded real speech, we can extract prosodic pattern about each phoneme, and assign them as target and test patterns for ANNs. ANNs can learn the prosody from natural speech and generate prosodic patterns of the central phonemic segment in phoneme strings as output response of ANNs when phoneme strings of a sentence are given to ANNs as input stimuli.

• 대한음성학회지:말소리
• /
• 제33_34호
• /
• pp.23-42
• /
• 1997

This study has two main purposes. One is to explore the relationship between syntactic aspects and prosodic aspects in Standard Korean. The other is to provide speech synthesis with the information about such relationship. This study will focus on the prosodic behavior of subjective particles'-i/-ga', '-eun/-neun'. The prosodic features of subjective particles are described respectively. How do the elements such as the position of particles in a sentence, the sentence constituents, the length of the sentence and the rhythmic boundaries influence on the prosodic behavior are also investigated.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제34권2호
• /
• pp.104-111
• /
• 2007

본 논문은 두 가지의 구조적 운율 정보, 즉 운율어와 음절수를 이용하여 한국어 위치 정보 데이타의 발음모델링을 수행할 경우에 음성인식기의 성능을 평가하는 것을 목표로 하는 이다. 먼저, 위치 정보 데이타가 운율어로 구성되어 있다는 전제 하에 운율어를 이용하여 위치 정보 데이타의 가능한 모든 발음을 생성하고, 다시 음절수를 기준으로 발음변이 수를 조절하는 방법을 제시하였다. 제안한 방법에 의하여 9개의 테스트 세트와 9개의 학습 세트로 총 81개의 실험을 통하여 음성인식의 성능을 평가하였다. 실험 결과 운율어를 이용하여 발음 사전을 제작한 모든 경우에 베이스라인과 비교하여 성능이 향상되었다. 음절수에 따라서 발음 변이의 수를 조절한 결과도 전체적으로는 3음절로 그 수를 제한한 경우에 가장 좋은 인식 성능을 얻을 수 있어서, 음절수에 따른 발음 변이 수의 조절이 효과적임을 알 수 있었다. 제안한 방법과 같이 운율어와 음절수를 이용한 경우에 베이스라인의 WER 4.63%에서 최대 8.4%의 WER가 감소하였다.

• 한국음향학회지
• /
• 제18권8호
• /
• pp.9-16
• /
• 1999

일반적으로 사람은 말을 할 때 어절들은 몇몇의 구로 그룹핑하여 발음함으로써 발화한다. 이것은 듣는 사람으로 하여금 발화의 의미와 의도를 잘 파악하도록 도와준다. 특히, 이러한 목적으로 발화자는 무의식적으로 운율정보(억양, 장단, 리듬 등)를 적절히 사용하게 된다. 본 논문에서는 발화된 문장에서 운율경계를 인식의 단위로 하는 음성인식방법에 대하여 제안한다. 즉, 발화된 문장을 운율구단위로 나누는 방법을 제안하고 나누어진 단위에 따라 연속음 인식실험을 수행하였다. 인식실험결과 연속음인식 시간의 감소를 관찰할 수 있었으며, 물론 음성인식률도 20-10%정도 증가하였다.

• 대한음성학회지:말소리
• /
• 제61호
• /
• pp.87-100
• /
• 2007

A punctuation generation system which combines prosodic information with acoustic and language model information is presented. Experiments have been conducted first for the reference text transcriptions. In these experiments, prosodic information was shown to be more useful than language model information. When these information sources are combined, an F-measure of up to 0.7830 was obtained for adding punctuation to a reference transcription. This method of punctuation generation can also be applied to the 1-best output of a speech recogniser. The 1-best output is first time aligned. Based on the time alignment information, prosodic features are generated. As in the approach applied in the punctuation generation for reference transcriptions, the best sequence of punctuation marks for this 1-best output is found using the prosodic feature model and an language model trained on texts which contain punctuation marks.

• 인지과학
• /
• 제23권2호
• /
• pp.249-267
• /
• 2012

본 연구에서는 구문 처리에서 운율 정보 활용이 작업 기억 용량의 영향을 받는지를 검증하였다. 구체적으로 작업 기억 용량이 운율 경계의 강도와 위치에 따른 관계절 부착 중의성 해소 방식 차이를 예측하는지를 알아보았다. 실험 결과, 작업 기억 폭이 큰 청자들의 중의성 해소 방식이 작업 기억 폭이 작은 청자들에 비해 운율 경계 강도의 영향을 더 받는 것으로 나타났다. 이는 다른 상위 수준 제약과 마찬가지로 운율 제약의 활용도 작업 기억과 같은 인지적 자원을 필요로 함을 시사한다.

• 말소리와 음성과학
• /
• 제15권3호
• /
• pp.1-7
• /
• 2023

This study investigates how the phonological and phonetic information of vowels influences prosodic prominence among linguistically untrained listeners using public speech in American English. We first examined the speech material's phonetic realization of vowels (i.e., maximum F0, F0 range, phone rate [as a measure of duration considering the speech rate of the utterance], and mean intensity). Results showed that the high vowels /i/ and /u/ likely had the highest max F0, while the low vowels /æ/ and /ɑ/ tended to have the highest mean intensity. Both high and low vowels had similarly high phone rates. Next, we examined the effects of the vowels' phonological and phonetic information on listeners' perceptions of prosodic prominence. The results showed that vowels significantly affected the likelihood of perceived prominence independent of acoustic cues. The high and low vowels affected probability of perceived prominence less than the mid vowels /ɛ/ and /ʌ/, although the former two were more likely to be phonetically enhanced in the speech than the latter. Overall, these results suggest that perceptions of prosodic prominence in English are not directly influenced by signal-driven factors (i.e., vowels' acoustic information) but are mediated by expectation-driven factors (e.g., vowels' phonological information).