Accent phrase segmentation by finding n-best sequences of pitch pattern templates
نویسندگان
چکیده
This paper describes a prosodic method for segmenting continuous speech into accent phrases. Optimum sequences are obtained on the basis of least squared error criterion by using dynamic time warping between F0 contours of input speech and reference accent patterns called ‘pitch pattern templates’. But the optimum sequence does not always give good agreement with phrase boundaries labeled by hand, while the second or the third optimum candidate sequence does well. Therefore, we expand our system to be able to find out multiple candidates by using N-best algorithm. Evaluation tests were carried out using the ATR continuous speech database of 10 speakers. The results showed about 97% of phrase boundaries were correctly detected when we took 30-best candidates, and this accuracy is 7.5% higher than the conventional method without using N-best search algorithm.
منابع مشابه
Prosodic phrase segmentation by pitch pattern clustering
This paper proposes a novel method for detect,ing the optimal sequence of prosodic phrases from continuous speech based on data-driven approach. The pitch pattern of input speech is divided into prosodic segments which minimized the overall distortion with pitch pattern templates of accent phrases by using the One Pass search algorithm. The pitch pattern templates are designed by clustering a l...
متن کاملAutomatic prosodic segmentation by F0 clustering using superpositional modeling
In this paper, we propose an automatic method for detecting accent phrase boundaries in Japanese continuous speech by using F0 information. In the training phase, hand labeled accent patterns are parameterized according to a superpositional model proposed by Fujisaki, and assigned to some clusters by a clustering method, in which accent templates are calculated as centroid of each cluster. In t...
متن کاملThe Structure of French Intonational Rises: A Study of Text-to-Tune Alignment
A production study examined the structure of French intonational rises. Prosodic phrases with a two-rise pattern (LHLH) were most common. Phrase length, expressed either in number of syllables or in clock time, was the best predictor of the realization of the two-rise pattern. Several other patterns were observed, including one not reported in the literature. I argue, following [6], that the ea...
متن کاملPhrase initial accent I in South Swedish
The topic of this paper is the variability of pitch realisation of phrase-initial accent I. In our study we have observed a difference in variability for the varieties investigated. Central Swedish pitch patterns for phrase-initial accent I both to the East (Stockholm) and to the West (Gothenburg) display an apparent constancy, albeit with distinct patterns: East Central Swedish rising and West...
متن کاملPitch Accent in Japanese: Implementation by the C/D Model
In Tokyo Japanese, lexical accent is implemented by pitch pattern control, while phrasal stress patterns, along with pitch variation, convey non-lexical information in discourse. The C/D model represents pitch control by the tonal melody and stress control by the skeletal organization of the utterance. Phonetic implementation of pitch contours is exemplified here for different lexical accent pa...
متن کامل