|
|
Automatic vocal-oriented recognition of human emotions
Houdek, Miroslav ; Přinosil, Jiří (referee) ; Atassi, Hicham (advisor)
This master thesis concerns with emotional states and gender recognition on the basis of speech signal analysis. We used various prosodic and cepstral features for the description of the speech signal. In the text we describe non-invasive methods for glottal pulses estimation. The described features of speech were implemented in MATLAB. For their classification we used the GMM classifier, which uses the Gaussian probability distribution for modeling a feature space. Furthermore, we constructed a system for recognition of emotional states of the speaker and a system for gender recognition from speech. We tested the success of created systems with several features on speech signal segments of various lengths and compared the results. In the last part we tested the influence of speaker and gender on the success of emotional states recognition.
|
|
|
The tempo of speech in theatrical Czech during two periods separated by several decades
Bartošová, Petra ; Palková, Zdenka (advisor) ; Skarnitzl, Radek (referee)
The diploma thesis deals with the issue of speech rate. The theoretical part briefly describes the investigation of speech tempo. It defines the types of tempo examined in this study (articulation rate and modified speaking rate) and factors that influence the tempo of speech. The practical part of the thesis aims to ascertain whether speech rate on the stage has increased and whether it is influenced by the type of text (monologue, dialogue, monological dialogue). The material consists of four theatre productions (Lakomec from 1972 and 2004 and Naši furianti from 1979 and 2006). Therefore we do not investigate gradual changes within the given periods, but instead compare two pairs of productions of the same dramatic text, realized with a time interval of approximately 30 years. A linear mixed-effects model was used as the main method for statistical evaluation of results gathered by measuring the speech rate. The results show that neither articulation rate nor modified speaking rate changed significantly in the observed productions. Differences with some statistical significance were obtained for comparison of tempo in texts of differing type, specifically in texts of different line lengths. The results relate especially to articulation rate, lesser to modified speaking rate.
|
|
|
Temporal characteristics in the speech of guides in repeated texts
Hanušová, Kateřina ; Palková, Zdena (advisor) ; Veroňková, Jitka (referee)
The guide's speech is a very specific type of discourse. It is a semi-planned monologue performed, with minimal deviations from the pattern, several times in succcession. These qualities therefore allow for a study of each speaker's articulatory rate. This paper analyzes the articulatory and speech rate, pauses, as well as the distinction of prosodic units, using four guides' speeches (utterances?). Each of the speeches was studied in two iterations. This paper presents a detailed description of the observed qualities for each speaker and a comprehensive summary comparison of the guides' speech patterns.
|
|
|
The tempo of speech in theatrical Czech during two periods separated by several decades
Bartošová, Petra ; Palková, Zdenka (advisor) ; Skarnitzl, Radek (referee)
The diploma thesis deals with the issue of speech rate. The theoretical part briefly describes the investigation of speech tempo. It defines the types of tempo examined in this study (articulation rate and modified speaking rate) and factors that influence the tempo of speech. The practical part of the thesis aims to ascertain whether speech rate on the stage has increased and whether it is influenced by the type of text (monologue, dialogue, monological dialogue). The material consists of four theatre productions (Lakomec from 1972 and 2004 and Naši furianti from 1979 and 2006). Therefore we do not investigate gradual changes within the given periods, but instead compare two pairs of productions of the same dramatic text, realized with a time interval of approximately 30 years. A linear mixed-effects model was used as the main method for statistical evaluation of results gathered by measuring the speech rate. The results show that neither articulation rate nor modified speaking rate changed significantly in the observed productions. Differences with some statistical significance were obtained for comparison of tempo in texts of differing type, specifically in texts of different line lengths. The results relate especially to articulation rate, lesser to modified speaking rate.
|
|
|
Speaker identification in the temporal domain of speech
Weingartová, Lenka ; Volín, Jan (advisor) ; Skarnitzl, Radek (referee) ; Pollák, Petr (referee)
This thesis aims to thoroughly describe the temporal characteristics of spoken Czech by means of phone durations and their changes under the influence of several prosodic and segmental factors, such as position in a higher unit (syllable, word or prosodic phrase), length of the higher unit, segmental environment, structure of the syllable or phrase-final lengthening. The speech material comes from a semi-spontaneous corpus of scripted dialogues comprising 4046 utterances by 34 speakers. The descriptions are afterwards used for the creation of a rule-based temporal model, which provides a baseline for analysing local articulation rate contours and their speaker-specificity. The results indicate, that systematic speaker-specific differences can be found in the segmental domain, as well as in the temporal contours. Moreover, speaker identification potential of articulation rate and global temporal features is also assessed. Keywords: temporal characteristics, temporal modelling, phone duration, speaker identification, Czech
|
|
|
Automatic vocal-oriented recognition of human emotions
Houdek, Miroslav ; Přinosil, Jiří (referee) ; Atassi, Hicham (advisor)
This master thesis concerns with emotional states and gender recognition on the basis of speech signal analysis. We used various prosodic and cepstral features for the description of the speech signal. In the text we describe non-invasive methods for glottal pulses estimation. The described features of speech were implemented in MATLAB. For their classification we used the GMM classifier, which uses the Gaussian probability distribution for modeling a feature space. Furthermore, we constructed a system for recognition of emotional states of the speaker and a system for gender recognition from speech. We tested the success of created systems with several features on speech signal segments of various lengths and compared the results. In the last part we tested the influence of speaker and gender on the success of emotional states recognition.
|
guest ::
