|
|
Emotion Recognition from Acted and Spontaneous Speech
Atassi, Hicham ; Přibil, Jiří (oponent) ; Zahradník, Pavel (oponent) ; Smékal, Zdeněk (vedoucí práce)
Doctoral thesis deals with emotion recognition from speech signals. The thesis is divided into two main parts; the first part describes proposed approaches for emotion recognition using two different multilingual databases of acted emotional speech. The main contributions of this part are detailed analysis of a big set of acoustic features, new classification schemes for vocal emotion recognition such as “emotion coupling” and new method for mapping discrete emotions into two-dimensional space. The second part of this thesis is devoted to emotion recognition using multilingual databases of spontaneous emotional speech, which is based on telephone records obtained from real call centers. The knowledge gained from experiments with emotion recognition from acted speech was exploited to design a new approach for classifying seven emotional states. The core of the proposed approach is a complex classification architecture based on the fusion of different systems. The thesis also examines the influence of speaker’s emotional state on gender recognition performance and proposes system for automatic identification of successful phone calls in call center by means of dialogue features.
|
|
|
Stress recognition from speech signal
Staněk, Miroslav ; Přibil, Jiří (oponent) ; Tučková,, Jana (oponent) ; Sigmund, Milan (vedoucí práce)
Presented doctoral thesis is focused on development of algorithms for psychological stress detection in speech signal. The novelty of this thesis aims on two different analysis of the speech signal- the analysis of vowel polygons and the analysis of glottal pulses. By performed experiments, the doctoral thesis uncovers the possible usage of both fundamental analyses for psychological stress detection in speech. The analysis of glottal pulses in amplitude domain according to Top-To-Bottom criterion seems to be as the most effective with the combination of properly chosen classifier, which can be defined as language and phoneme independent way to stress recognition. All experiments were performed on developed Czech real stress database and some observations were also made on English database SUSAS. The variety of possibly effective ways of stress recognition in speech leads to approach very high recognition accuracy of their combination, or of their possible usage for detection of other speaker’s state, which has to be further tested and verified by appropriate databases.
|
|
|
Biologické změny CNS po traumatech
Maulenová, Kristýna ; Šivicová, Gabriela (vedoucí práce) ; Kulišťák, Petr (oponent)
Tato práce se zaměřuje na traumatická poranění mozku a jejich negativní dopad na mozek jako celek i jednotlivé mozkové struktury. První dvě kapitoly teoretické části shrnují informace o anatomii a funkcích centrální nervové soustavy a nejčastější příčiny traumat a jejich charakteristiku. Stěžejní bod práce tvoří psychické následky traumat, jež jsou členěny dle mozkového laloku, který se hlavní měrou podílí na klinickém projevu poškozených funkcí. Práce dále shrnuje oblasti neurodiagnostiky a uvádí některé konkrétní metody využívané při práci s pacienty po traumatech. Poslední kapitola teoretické části se zabývá možnostmi a příklady neurorehabilitace i faktory, jež se podílí na prognóze posttraumatického stavu. Část této kapitoly se rovněž věnuje plasticitě mozku. Poslední dvě kapitoly odkazují na další doporučenou literaturu. Empirická část podává návrh výzkumu změn zapojených mozkových částí u pacientů po traumatických poraněních mozku při rozpoznávání emocí vyjádřených obličejem pomocí Ekmanova testu rozpoznání emocí a zobrazovací metodou SPECT. Klíčová slova: Centrální nervová soustava, mozkové laloky, neurodiagnostika, psychické následky, rozpoznání emocí, traumatické poranění mozku
|
|
|
Stress recognition from speech signal
Staněk, Miroslav ; Přibil, Jiří (oponent) ; Tučková,, Jana (oponent) ; Sigmund, Milan (vedoucí práce)
Presented doctoral thesis is focused on development of algorithms for psychological stress detection in speech signal. The novelty of this thesis aims on two different analysis of the speech signal- the analysis of vowel polygons and the analysis of glottal pulses. By performed experiments, the doctoral thesis uncovers the possible usage of both fundamental analyses for psychological stress detection in speech. The analysis of glottal pulses in amplitude domain according to Top-To-Bottom criterion seems to be as the most effective with the combination of properly chosen classifier, which can be defined as language and phoneme independent way to stress recognition. All experiments were performed on developed Czech real stress database and some observations were also made on English database SUSAS. The variety of possibly effective ways of stress recognition in speech leads to approach very high recognition accuracy of their combination, or of their possible usage for detection of other speaker’s state, which has to be further tested and verified by appropriate databases.
|
|
|
Emotion Recognition from Acted and Spontaneous Speech
Atassi, Hicham ; Přibil, Jiří (oponent) ; Zahradník, Pavel (oponent) ; Smékal, Zdeněk (vedoucí práce)
Doctoral thesis deals with emotion recognition from speech signals. The thesis is divided into two main parts; the first part describes proposed approaches for emotion recognition using two different multilingual databases of acted emotional speech. The main contributions of this part are detailed analysis of a big set of acoustic features, new classification schemes for vocal emotion recognition such as “emotion coupling” and new method for mapping discrete emotions into two-dimensional space. The second part of this thesis is devoted to emotion recognition using multilingual databases of spontaneous emotional speech, which is based on telephone records obtained from real call centers. The knowledge gained from experiments with emotion recognition from acted speech was exploited to design a new approach for classifying seven emotional states. The core of the proposed approach is a complex classification architecture based on the fusion of different systems. The thesis also examines the influence of speaker’s emotional state on gender recognition performance and proposes system for automatic identification of successful phone calls in call center by means of dialogue features.
|
host ::
RSS.