Perception of stress and speaking style for selected elements of the SUSAS database

Bolia, Robert S.; Slyh, Raymond E.

doi:10.1016/s0167-6393(02)00129-2

Cited by 8 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The first data collection is Danish Emotion Speech (DES) containing recordings of speech utterances expressed by 4 actors in 5 emotional states [13]. The second data collection uses a subset of Speech Under Simulated and Actual Stress (SUSAS) data collection which includes words uttered under low and high stress conditions as well as speech in various talking styles expressed by 9 native American English speakers [14,15]. Several statistics of pitch, formants, and energy contours were extracted as features [16].…”

Section: Emotional Speech Data Collectionsmentioning

confidence: 99%

Feature Selection Based on Mutual Correlation

Haindl

Somol

Ververidis

et al. 2006

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Feature selection is a critical procedure in many pattern recognition applications. There are two distinct mechanisms for feature selection namely the wrapper methods and the filter methods. The filter methods are generally considered inferior to wrapper methods, however wrapper methods are computationally more demanding than filter methods. A novel filter feature selection method based on mutual correlation is proposed. We assess the classification performance of the proposed filter method by using the selected features to the Bayes classifier. Alternative filter feature selection methods that optimize either the Bhattacharyya distance or the divergence are also tested. Furthermore, wrapper feature selection techniques employing several search strategies such as the sequential forward search, the oscillating search, and the sequential floating forward search are also included in the comparative study. A trade off between the classification accuracy and the feature set dimensionality is demonstrated on both two benchmark datasets from UCI repository and two emotional speech data collections.

show abstract

Section: Emotional Speech Data Collectionsmentioning

confidence: 99%

Feature Selection Based on Mutual Correlation

Haindl

Somol

Ververidis

et al. 2006

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…Gussenhoven's study [9] of stress shift shew that stress shift as a rhetorical device was particularly frequent in propagandist speech, as used for instance in commercials and political speeches. Bolia and Slyh [10] worked on the relationship between perception of stress and speaking styles in monosyllabic or disyllabic words, and their result shew listeners correctly classified the utterance 58% of the time. Chu and Bao [11] recently studied stress distribution tendencies within base phrases among different reading styles.…”

Section: Introductionmentioning

confidence: 99%

A comparative study of sentential stress distribution in Mandarin multi-style speeches

Bao,

Chu

2006

Speech Prosody 2006

View full text Add to dashboard Cite

This paper compares the distribution of sentential stresses among three speaking styles: Lyric, Critical, and Explanatory; and extends our previous study in the base phrase level to the sentence construction level and the prosodic word level. The results show that 1) The distributions of both rhythmic and semantic stresses act the same among styles within prosodic words, although the distribution tendencies change due to different structure properties of the words; 2) In the sentence construction level, the distribution tendency of rhythmic stress is quite similar across three styles in most construction types, while semantic stress presents more diversity among speaking styles. The Explanatory style shares a similar tendency with the Neutral style. The Lyric style differs from the Neutral style in constructions with the subject-predicate structure; the Critical style differs in constructions with the predicate-object, the adjunct-subject, and the adjunct-object structures. Generally, speaking styles have fewer effects on rhythmic stress distribution than on semantic stress. Such effects are more obvious in the sentence construction and the base phrase levels than the prosodic word level, where syntax plays a more crucial role in stress distribution.

show abstract

Military Applications: Human Factors Aspects of Speech-Based Systems

Noyes

Haas

2010

Speech Technology

View full text Add to dashboard Cite

Perception of stress and speaking style for selected elements of the SUSAS database

Cited by 8 publications

References 16 publications

Feature Selection Based on Mutual Correlation

Feature Selection Based on Mutual Correlation

A comparative study of sentential stress distribution in Mandarin multi-style speeches

Military Applications: Human Factors Aspects of Speech-Based Systems

Contact Info

Product

Resources

About