|
|
The relation of emotions and intonation curves
Gavlasová, Radka ; Smékal, Zdeněk (referee) ; Tučková,, Jana (advisor)
This thesis deals with intonation curves and their relation to human emotions. Besides the theoretical part where you can learn about speech production, signal processing and psychological distribution of emotions, there is also a unique database recorded with the help of two professional actors. The main goal of this thesis is to classify created data using artificial neural networks into four classes. Those classes are anger, joy, boredom and sadness. The practical part was implemented in a programming platform called Matlab using Classification Learner app. Features used for this method were variations of fundamental frequency and MFCC. The results were compared with a listening survey so that it could be determined whether the results provided by neural network are relevant to some kind of a human factor. Success rate of the trained models reached 82 %, new data testing reached 75 %. Listening survey confirmed that the results correspond to the assumption of human perception. Better success rate would be accomplished by using a bigger set of higher quality data.
|
|
|
Voice and its modulation in human mate choice
Vobecká, Tereza ; Třebický, Vít (advisor) ; Schneiderová, Irena (referee)
1 Abstract Vocal communication is one of the primary means of communication in humans. Information from utterances not providedsolely by speech but also by their acoustic properties (e.g. voice pitch). People tend to associate different acoustic parameters of voice with specific human characteristics, such as physical traits such as body size, sex or personality traits such as dominance or attractiveness. Human voice is dynamic, and it is modulated depending on the inner state of the speaker and external circumstances. For example, the modulation of speaker's voice may be caused by their emotional state or depend on the person they are talking to. This work aims to summarize and review existing literature about the modulation of acoustic parameters of the voice and its effect on perception of others in mate-choice context. Key words acoustics, attractiveness, non-verbal vocalization, fundamental frequency, preferences
|
|
|
Detecting Stress in Speech
Šoltés, Samuel ; Beneš, Karel (referee) ; Grézl, František (advisor)
Stress influences people in several ways and can lead to decrease in performance and / or critical mistakes. Stress detection in speech measures the influence of stress in speech. The goal of this thesis is to offer a closer look at the impacts of stress, choose adequate parameters of speech which would manifest these impacts, implement their estimation and compare their results. The thesis contains description of stress and its effects on humans; glottal pulse, spectrum, fundamental frequency and formants as the parameters chosen for stress estimation; design and implementation of parameter value estimation from speech signal and obtained values of given parameters on two different databases.
|
|
|
The nuclear accent instantiation in English texts read by Czech speakers
Vavryčuk, Václav ; Volín, Jan (advisor) ; Studenovský, David (referee)
The thesis is concerned with Czech accent in English in the intonation domain, specifically the differences between nuclear melodic patterns in read speech of Czech speakers with a heavy accent and native British speakers. It is aiming to contribute to the study of "Czech" English with a later possibility of using the results in language teaching. The first part focuses on foreign accent in general together with its connection to comprehensibility and the influence of the first language. We then go through some theoretical concepts in studying intonation and their application in Czech and English intonology. In the experimental part we compare the melodic patterns of Czech and British speakers using cluster analysis. The main differences are evident in non-final utterances, where Czech speakers use mostly rising contours while British speakers more often use falling contours.
|
|
|
Ways of exploiting fundamental frequency for speaker identification
Hývlová, Dita ; Skarnitzl, Radek (advisor) ; Šturm, Pavel (referee)
The present Master's thesis deals with the forensic use of fundamental frequency characteristics, specifically with F0 mean values and indicators of variability. Phoneticians who specialise in the forensic analysis of speech generally believe that F0 does not hold much potential as a parameter useful for speaker identification, mainly because it is easily influenced by extrinsic factors (e.g. the speaker's emotional state, interfering noise, transmission channel or even the speaker's own effort to mask his voice), which cause high intra-individual variability. Despite these facts, however, the forensic use of F0 offers a number of advantages, namely straightforward extraction from the speech signal and lower susceptibility to varying lexical content - unlike, for example, vowel formants. This thesis investigates the recordings of 8 male speakers made in two different speech styles (spontaneous and read) and compares the respective indicators of F0 stability and variability, in particular those that are robust in varying external conditions: that is, the baseline for mean values and the 10.-90. percentile range for variability indicators. Apart from that, we take into account phenomena such as the creaky voice, which are idiosyncratic and contribute to easier speaker discrimination. Key words:...
|
|
|
Perceptual sensitivity to music and speech stimuli in the frequency and temporal domains
Lukeš, David ; Volín, Jan (advisor) ; Skarnitzl, Radek (referee)
The subject of this thesis is perceptual sensitivity with respect to subtle frequency-based and temporal manipulations in speech, music and mixed stimuli. We hypothesize that an individual's sensitivity to variation in all three types of stimuli should be similar (i.e. a correlation should exist), seeing that findings in evolutionary biology, neurosciences, psy- chology and experimental phonetics are pointing towards a relatively strong link between the mechanisms of perception in speech and music. Our listening experiment revealed mostly intermediate correlations; additionally, we argue that by employing syntactically less complicated stimuli, which would target specifically fundamental sensitivity without requiring a complex syntactic analysis in parallel, even more robust correlations could be obtained. While the influence of prior formal linguistic education on performance in the test was negligible, the influence of musical experience was considerable, which lends further support to the idea of simplifying especially the music stimuli in future research. Key words: music, speech, perception, sensitivity, correlation
|
|
|
The just noticeable difference for English melodic prominence elicited on Czech listeners
Brabcová, Kateřina ; Šturm, Pavel (advisor) ; Tichý, Ondřej (referee)
The aim of the thesis is to ascertain whether and how the modification of pitch in the acoustic signal influences what we hear, i.e. whether we perceive a difference between the original, unmodified stimulus and the one with a modified melodic contour. The theoretical part of the thesis explains terms such as: the fundamental frequency and its contour, the just noticeable difference, intonation and the structure of intonational phrases, and the difference between English and Czech intonational structure. The practical part uses recordings of Standard British English. The contour of the fundamental frequency (F0) is changed at different places of the intonation phrase (the nuclear syllable, stressed and unstressed syllables of the head) and the pitch is either lowered or raised by 1.5 semitones with respect to the original production. The main goal is (1) to determine the influence of the melodic prominence of a syllable on the pitch manipulation detection (judged by 20 listeners of Czech origin, all students of English) and (2) to find out which other factors (such as the already mentioned direction of the change) might be significant. The results are analysed and their statistical significance is evaluated.
|
|
|
Speech signal processing in time domain
Marko, Ján ; Staněk, Miroslav (referee) ; Sigmund, Milan (advisor)
The bachelor´s project deals with the processing of speech signals. The work includes a search of available publications on the issue of determining the voiced segments of speech and fundamental frequency. Attention is devoted to methods for speech recognition features in the time domain. Comparing theoretical and practical results on the real speech signals evaluates the use of methods, highlighting their advantages and disadvantages.
|
|
|
Detection of the voice fundamental frequency
Chloupek, Jiří ; Mekyska, Jiří (referee) ; Sysel, Petr (advisor)
This bachelor thesis deals with the detection of the pitch man. The frequency of the basic tone is one of the basic parameters of speech signal in the frequency domain. In this thesis we describe several methods for pitch detection and practical application of correlation method and cepstral analysis.
|
|
|
Real-Time Analysis of Audio Signals
Řezáč, Martin ; Schimmel, Jiří (referee) ; Černocký, Jan (advisor)
The goal of this thesis is creation an application, which can perform real-time fundamental frequency tracking of incoming audio samples. According to detected frequencies, the program generates MIDI messages, which are sent to chosen MIDI device. First, the reader is introduced to the issue of fundamental frequency tracking. The following part describes individual methods, especially the one based on spectral analysis of a tone. A description of used technologies is also a included in this part of this thesis. In the following part, the implementation and testing of application are described including opinions of several musicians about this product. At the end, the whole work is concluded and the possible further development is outlined.
|
guest ::
