Modeling stylized invariance and local variability of prosody in text-to-speech synthesis

Chen, Min; Zhao, Yong; Chang, Eric

doi:10.1016/j.specom.2005.10.003

Cited by 11 publications

(10 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the basic characteristics of natural speech is its variability ͑Braun et al, Chu et al, 2006͒. In order to generate models for speech production, synthesis, and perception this variability should be accounted for appropriately.…”

Section: Emotional Regions In F0 Mean-range Spacementioning

confidence: 99%

“…For example, its implementation in emotional speech synthesis is limited ͑Cowie et al, 2001͒ because it does not specifically account for the variability present in the natural speech ͑Braun et Chu et al, 2006;Pell, 2001͒. In the traditional analysis, an emotional utterance is represented as a point in the parameter space. We suggest a new model where each utterance is represented by an "emotional region" in the parameter space.…”

Section: Introductionmentioning

confidence: 99%

“…The concept of the variability of prosodic patterns was studied in a database composed of two repetitions of 1000 sentences recorded with six months separation by Chu et al ͑2006͒. The results showed wide variations in F0 values, sometimes corresponding to 50% of the dynamical range of the speaker.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On the robustness of overall F0-only modifications to the perception of emotions in speech

Bulut

Narayanan

2008

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

Emotional information in speech is commonly described in terms of prosody features such as F0, duration, and energy. In this paper, the focus is on how F0 characteristics can be used to effectively parametrize emotional quality in speech signals. Using an analysis-by-synthesis approach, F0 mean, range, and shape properties of emotional utterances are systematically modified. The results show the aspects of the F0 parameter that can be modified without causing any significant changes in the perception of emotions. To model this behavior the concept of emotional regions is introduced. Emotional regions represent the variability present in the emotional speech and provide a new procedure for studying speech cues for judgments of emotion. The method is applied to F0 but can be also used on other aspects of prosody such as duration or loudness. Statistical analysis of the factors affecting the emotional regions, and discussion of the effects of F0 modifications on the emotion and speech quality perception are also presented. The results show that F0 range is more important than F0 mean for emotion expression.

show abstract

Section: Emotional Regions In F0 Mean-range Spacementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

On the robustness of overall F0-only modifications to the perception of emotions in speech

Bulut

Narayanan

2008

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

show abstract

“…Because different pronunciations of the same word with a good prosody can result in very different prosodic patterns [12], it is necessary to have several gold standard references at our disposal when it comes to comparing prosodic contours in the task of prosody assessment. We chose to use the k-means algorithm to generate several reference contours by clustering similar prosodic contours together, as suggested by Cheng [7].…”

Section: Gold-standard Native Referencesmentioning

confidence: 99%

Automatic Assessment of L2 English Word Prosody Using Weighted Distances of F0 and Intensity Contours

2018

View full text Add to dashboard Cite

In the current paper, an automatic prosody assessment method for learners of English using a weighted comparison of fundamental frequency (F0) and intensity contours is proposed. Patterns of F0 and intensity of learners are compared to that of native using a proposed metric -a weighted distance -in which the error around the high values of prosodic features have more weight in the computation of the final distance. Gold-standard native references are built using the k-means clustering algorithm. Therefore, we also propose a data-driven criterion called weighted variance based on the weighted similarity within the whole set of native utterances to determine the optimal number of clusters k. In comparison with baseline contour comparison metrics which resulted in a subjective-objective score correlation of 0.278, our method combining the proposed metric and criterion led to a final subjective-objective score correlation of 0.304. In comparison, subjective scores correlated at 0.480.

show abstract

“…When different speakers are required to read the same sentence, prosodic diversity is usually observed. This diversity involves pitch accents placement and shape, phrase breaks location and other aspects of prosody, and it is even observed when it is a single speaker who reads the same sentence at two different times [2]. The variability observed in pitch accents and phrase breaks placement has suggested a distinction between optional and compulsory prosodic events, a distinction that, in turn, has been used to reformulate the evaluation of automatic predictors ( [3], [4], [5]).…”

Section: Introductionmentioning

confidence: 99%

Including pitch accent optionality in unit selection text-to-speech synthesis

Badino¹,

Clark²,

Strom³

2008

Interspeech 2008

View full text Add to dashboard Cite

A significant variability in pitch accent placement is found when comparing the patterns of prosodic prominence realized by different English speakers reading the same sentences. In this paper we describe a simple approach to incorporate this variability to synthesize prosodic prominence in unit selection text-to-speech synthesis. The main motivation of our approach is that by taking into account the variability of accent placements we enlarge the set of prosodically acceptable speech units, thus increasing the chances of selecting a good quality sequence of units, both in prosodic and segmental terms. Results on a large scale perceptual test show the benefits of our approach and indicate directions for further improvements.

show abstract

Modeling stylized invariance and local variability of prosody in text-to-speech synthesis

Cited by 11 publications

References 25 publications

On the robustness of overall F0-only modifications to the perception of emotions in speech

On the robustness of overall F0-only modifications to the perception of emotions in speech

Automatic Assessment of L2 English Word Prosody Using Weighted Distances of F0 and Intensity Contours

Including pitch accent optionality in unit selection text-to-speech synthesis

Contact Info

Product

Resources

About