Articulatory features describe how articulators are involved in making sounds.Speakers often use a more exaggerated way to pronounce accented phonemes,so articulatory features can be helpful in pitch accent detection....Articulatory features describe how articulators are involved in making sounds.Speakers often use a more exaggerated way to pronounce accented phonemes,so articulatory features can be helpful in pitch accent detection.Instead of using the actual articulatory features obtained by direct measurement of articulators,we use the posterior probabilities produced by multi-layer perceptrons(MLPs) as articulatory features.The inputs of MLPs are frame-level acoustic features pre-processed using the split temporal context-2(STC-2) approach.The outputs are the posterior probabilities of a set of articulatory attributes.These posterior probabilities are averaged piecewise within the range of syllables and eventually act as syllable-level articulatory features.This work is the first to introduce articulatory features into pitch accent detection.Using the articulatory features extracted in this way,together with other traditional acoustic features,can improve the accuracy of pitch accent detection by about 2%.展开更多
To investigate how a low tone （tone-3, T3） syllable in Chinese can be perceived to be focal accented or not, a total of 156 sentences containing tone-3 words were synthesized and used as stimuli in a perceptual stud...To investigate how a low tone （tone-3, T3） syllable in Chinese can be perceived to be focal accented or not, a total of 156 sentences containing tone-3 words were synthesized and used as stimuli in a perceptual study. The sentences differed in the falling value between the two high pitches, and in the duration and phonation types of the T3 syllables. Thirty-nine subjects were asked to judge where the focus or accent was for each sentence. The results show that at least three degrees of pitch drop are involved in the focus recognition： a big sized drop of about 10 semitones; a middle sized drop of about 6 semitones; a small sized drop of about 2 semitones. The results suggest that the three sizes of pitch drop have different indications in Chinese intonation, depending on both the tone and the tone combination. In perception, there are various ways to realize tone-3 focus in the Tx-T3-Ty sentences series, but in production or for text-to-speech synthesis, the rule simply is making a middle sized pitch drop with a long and creaky T3 syllable. Similarly, to focus on the low tone syllable in the T3-Tx-Ty sentences, a creaky T3 syllable is essential. However, a long T3 syllable is a strong determinant for a low tone focus in the Tx-Ty-T3 sentences.展开更多
基金Project(Nos.61370034,61273268,and 61005019) supported by the National Natural Science Foundation of China
文摘Articulatory features describe how articulators are involved in making sounds.Speakers often use a more exaggerated way to pronounce accented phonemes,so articulatory features can be helpful in pitch accent detection.Instead of using the actual articulatory features obtained by direct measurement of articulators,we use the posterior probabilities produced by multi-layer perceptrons(MLPs) as articulatory features.The inputs of MLPs are frame-level acoustic features pre-processed using the split temporal context-2(STC-2) approach.The outputs are the posterior probabilities of a set of articulatory attributes.These posterior probabilities are averaged piecewise within the range of syllables and eventually act as syllable-level articulatory features.This work is the first to introduce articulatory features into pitch accent detection.Using the articulatory features extracted in this way,together with other traditional acoustic features,can improve the accuracy of pitch accent detection by about 2%.
基金the National Office for Teaching Chinese as a Foreign Language (No. HBK01-05/17)
文摘To investigate how a low tone （tone-3, T3） syllable in Chinese can be perceived to be focal accented or not, a total of 156 sentences containing tone-3 words were synthesized and used as stimuli in a perceptual study. The sentences differed in the falling value between the two high pitches, and in the duration and phonation types of the T3 syllables. Thirty-nine subjects were asked to judge where the focus or accent was for each sentence. The results show that at least three degrees of pitch drop are involved in the focus recognition： a big sized drop of about 10 semitones; a middle sized drop of about 6 semitones; a small sized drop of about 2 semitones. The results suggest that the three sizes of pitch drop have different indications in Chinese intonation, depending on both the tone and the tone combination. In perception, there are various ways to realize tone-3 focus in the Tx-T3-Ty sentences series, but in production or for text-to-speech synthesis, the rule simply is making a middle sized pitch drop with a long and creaky T3 syllable. Similarly, to focus on the low tone syllable in the T3-Tx-Ty sentences, a creaky T3 syllable is essential. However, a long T3 syllable is a strong determinant for a low tone focus in the Tx-Ty-T3 sentences.
基金IRG:Internal Research Grants“The Study of Chinese Personal Pronouns from Typological Perspective”(RG 53/2018-2019R)Start-up Grants“The study of tone,intonation and SFP from typological perspective”(RG 35/2018-2019R)。
