<p>U okviru ove disertacije razvijeno je više različitih modela trajanja glasova u srpskom jeziku primenom odgovarajućih metoda automatskog učenja. Izvršena je objektivna evaluacija razvijenih modela i njihovo međusobno poređenje na osnovu kvantitativnih pokazatelja kao što su RMSE(engl. root-mean-squared error), MAE (engl. mean absolute error) i CC (engl. correlation coefficient). Takođe je izvršeno poređenje modela za srpski jezik sa performansama modela razvijenih za druge jezike, pri čemu je uočeno da su performanse modela razvijenih u ovoj disertaciji uporedljive ili čak prevazilaze performanse modela koji su razvijeni za druge jezike.</p> / <p>In this dissertation several different phone duration models of the Serbain<br />language using appropriate machine learning algorithms were developed.<br />The objective evaluation of the models obtained and their mutual comparison<br />based on quantitative measures such as RMSE (root-mean-squared error),<br />MAE (mean absolute error) and CC (correlation coefficient) were performed.<br />The comparison of the models developed for the Serbian language with the<br />performances of the models developed for other languages is also carried<br />out. It was observed that the performances of the models developed in this<br />dissertation are comparable or even outperform the performances of the<br />models that have been developed for other languages.</p>

Two heretofore unanalyzed aspects are addressed in this dissertation: 1. Building a model capable of predicting phone duration of Lithuanian. All existing investigations of phone durations of Lithuanian were performed by linguists. Usually these investigations are the kind of exploratory statistics and are limited to a single factor, affecting phone duration, analysis. Phone duration dependencies on contextual factors were estimated and written in explicit form (decision tree) in this work by means of machine learning method. 2. Construction of language independent method for creating phone duration models using very large, multi-speaker, automatically annotated speech corpus. Most of the researchers worldwide use speech corpus that are: relatively small scale, single speaker, manually annotated or at least validated by experts. Usually the referred reasons are: using multi-speaker speech corpora is inappropriate because different speakers have different pronunciation manners and speak in different speech rate; automatically annotated corpuses lack accuracy. The created method for phone duration modeling enables the use of such corpus. The main components of the created method are: the reduction of noisy data in speech corpus; normalization of speaker specific phone durations by using phone type clustering. The performed listening tests of synthesized speech, showed that: the perceived naturalness is affected by the underlying phones durations; The use of contextual... [to full text] / Disertacijoje nagrinėjamos dvi iki šiol netyrinėtos problemos: 1. Lietuvių kalbos garsų trukmių prognozavimo modelių kūrimas Iki šiol visi darbai, kuriuose yra nagrinėjamos lietuvių kalbos garsų trukmės, yra atlikti kalbininkų, tačiau šie tyrimai yra daugiau aprašomosios statistikos pobūdžio ir apsiriboja pavienių požymių įtakos garso trukmei analize. Šiame darbe, mašininio mokymo algoritmo pagalba, požymių įtaka garsų trukmei yra išmokstama iš duomenų ir užrašoma sprendimo medžio pavidalu. 2. Nuo kalbos nepriklausomų garsų trukmių prognozavimo modelių kūrimo metodas, naudojant didelės apimties daugelio, kalbėtojų automatiškai, anotuotą garsyną. Dėl skirtingų kalbėtojų tarties specifikos ir dėl automatinio anotavimo netikslumų, kuriant garsų trukmės modelius visame pasaulyje yra apsiribojama vieno kalbėtojo ekspertų anotuotais nedidelės apimties garsynais. Darbe pasiūlyti skirtingų kalbėtojų tarties ypatybių normalizavimo ir garsyno duomenų triukšmo atmetimo algoritmai leidžia garsų trukmių modelių kūrimui naudoti didelės apimties, daugelio kalbėtojų automatiškai anotuotus garsynus. Darbo metu atliktas audicinis tyrimas, kurio pagalba parodoma, kad šnekos signalą sudarančių garsų trukmės turi įtakos klausytojų/respondentų suvokiamam šnekos signalo natūralumui; kontekstinės informacijos panaudojimas garsų trukmių prognozavimo uždavinio sprendime yra svarbus faktorius įtakojantis sintezuotos šnekos natūralumą; natūralaus šnekos signalo atžvilgiu, geriausiai vertinamas yra... [toliau žr. visą tekstą]

The term prosody refers to characteristics of speech such as intonation, timing, loudness, and other acoustical properties imposed by physical, intentional and emotional state of the speaker. Phone durations and fundamental frequency contours are considered as two of the most prominent aspects of prosody. Modeling phone durations and fundamental frequency contours in Turkish speech are studied in this thesis. Various methods exist for building prosody models. State-of-the-art is dominated by corpus-based methods. This study introduces corpus-based approaches using classification and regression trees to discover the relationships between prosodic attributes and phone durations or fundamental frequency contours. In this context, a speech corpus, designed to have specific phonetic and prosodic content has been recorded and annotated. A set of prosodic attributes are compiled. The elements of the set are determined based on linguistic studies and literature surveys. The relevances of prosodic attributes are investigated by statistical measures such as mutual information and information gain. Fundamental frequency contour and phone duration modeling are handled as independent problems. Phone durations are predicted by using regression trees where the set of prosodic attributes is formed by forward selection. Quantization of phone durations is studied to improve prediction quality. A two-stage duration prediction process is proposed for handling specific ranges of phone duration values. Scaling and shifting of predicted durations are proposed to minimize mean squared error. Fundamental frequency contour modeling is studied under two different frameworks. One of them generates a codebook of syllable-fundamental-frequency-contours by vector quantization. The codewords are used to predict sentence fundamental frequency contours. Pitch accent prediction by two different clustering of codewords into accented and not-accented subsets is also considered in this framework. Based on the experience, the other approach is initiated. An algorithm has been developed to identify syllables having perceptual prominence or pitch accents. The slope of fundamental frequency contours are then predicted for the syllables identified as accented. Pitch contours of sentences are predicted using the duration information and estimated slope values. Performance of the phone duration and fundamental frequency contour models are evaluated quantitatively using statistical measures such as mean absolute error, root mean squared error, correlation and by kappa coefficients, and by correct classification rate in case of discrete symbol prediction.

Higher-level features are considered to be a potential remedy against transmission line and cross-channel degradations, currently some of the biggest problems associated with speaker verification. Phoneme durations in particular are not altered by these factors; thus a robust duration model will be a particularly useful addition to traditional cepstral based speaker verification systems. In this dissertation we investigate the feasibility of phoneme durations as a feature for speaker verification. Simple speaker specific triphone duration models are created to statistically represent the phoneme durations. Durations are obtained from an automatic hidden Markov model (HMM) based automatic speech recognition system and are modeled using single mixture Gaussian distributions. These models are applied in a speaker verification system (trained and tested on the YOHO corpus) and found to be a useful feature, even when used in isolation. When fused with acoustic features, verification performance increases significantly. A novel speech rate normalization technique is developed in order to remove some of the inherent intra-speaker variability (due to differing speech rates). Speech rate variability has a negative impact on both speaker verification and automatic speech recognition. Although the duration modelling seems to benefit only slightly from this procedure, the fused system performance improvement is substantial. Other factors known to influence the duration of phonemes are incorporated into the duration model. Utterance final lengthening is known be a consistent effect and thus “position in sentence” is modeled. “Position in word” is also modeled since triphones do not provide enough contextual information. This is found to improve performance since some vowels’ duration are particularly sensitive to its position in the word. Data scarcity becomes a problem when building speaker specific duration models. By using information from available data, unknown durations can be predicted in an attempt to overcome the data scarcity problem. To this end we develop a novel approach to predict unknown phoneme durations from the values of known phoneme durations for a particular speaker, based on the maximum likelihood criterion. This model is based on the observation that phonemes from the same broad phonetic class tend to co-vary strongly, but that there is also significant cross-class correlations. This approach is tested on the TIMIT corpus and found to be more accurate than using back-off techniques. / Dissertation (MEng)--University of Pretoria, 2009. / Electrical, Electronic and Computer Engineering / unrestricted