Building Text Corpus for Unit Selection Synthesis

Kasparaitis, Pijus; Anbinderis, T.

doi:10.15388/informatica.2014.29

Cited by 4 publications

(9 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Lithuanian unit selection synthesizer presented in (Kasparaitis and Anbinderis, 2014) will be used in the present work. It was created based on the proposal made by Yi and Glass (2002) to calculate the costs between phonological classes rather than between the sounds of speech.…”

Section: Phoneme Based Synthesizermentioning

confidence: 99%

“…The base type of this synthesizer is phoneme. The following improvements have been made as compared with the synthesizer described in (Kasparaitis and Anbinderis, 2014): a) New joining and substitution costs were calculated. Before calculating the costs research was carried out (Kasparaitis and Skersys, 2015) seeking to find the best set of parameters and the distance metric so that the distance between the instances of the same phoneme were as small as possible, and the distance between the instances of different phonemes were as large as possible.…”

Section: Phoneme Based Synthesizermentioning

confidence: 99%

“…The optimized version of the Viterbi search algorithm described by Kasparaitis and Anbinderis (2014) was used in the phoneme based synthesizer. Since this optimization is impossible in some types of diphone synthesizers, the Viterbi search algorithm was programmed anew in diphone synthesizers.…”

Section: Diphone Based Synthesizersmentioning

confidence: 99%

“…The algorithm for selecting the sentences to be recorded is described by Kasparaitis and Anbinderis (2014). A set containing 5000 sentences having the best coverage of 3-phones and 4-phones of Lithuanian was created.…”

Section: Sound Databasementioning

confidence: 99%

“…A remark should be made that the phonemic synthesizer was optimized as described in (Kasparaitis and Anbinderis, 2014) whereas both versions of diphone synthesizers were run without being optimized. The number of combinations, which the un-optimized phonemic and diphone synthesizers had to check in order to find the best combination of synthesis units, was calculated using the same 20000 sentences.…”

Section: Performance Of Algorithmsmentioning

confidence: 99%

See 4 more Smart Citations

Phoneme vs. Diphone in Unit Selection TTS of Lithuanian

Kasparaitis¹,

Kancys²

2018

BJMC

Self Cite

View full text Add to dashboard Cite

Abstract. The present paper deals with choosing the base type for the unit selection speech synthesis method of the Lithuanian language. Phoneme and diphone units have been examined. Besides, two different methods of joining costs calculation were employed in a diphone synthesizer: one was based on the spectral similarity and the other was based on phonological classes of the sounds to be joined. Synthesizers were evaluated according to their performance, algorithm complexity, the number of joins in a synthesized speech and the human listeners' subjective judgment. Experimental testing showed that the diphone synthesizer based on phonological classes was much more acceptable to the listeners than the one based on the spectral similarity. The diphone synthesizer based on phonological classes outperformed the phonemic synthesizer in terms of performance and the number of joins though it was somewhat less acceptable to human listeners.

show abstract

Section: Phoneme Based Synthesizermentioning

confidence: 99%

Section: Phoneme Based Synthesizermentioning

confidence: 99%

Section: Diphone Based Synthesizersmentioning

confidence: 99%

Section: Sound Databasementioning

confidence: 99%

Section: Performance Of Algorithmsmentioning

confidence: 99%

See 3 more Smart Citations

Phoneme vs. Diphone in Unit Selection TTS of Lithuanian

Kasparaitis¹,

Kancys²

2018

BJMC

Self Cite

View full text Add to dashboard Cite

show abstract

Automatic Parameters Estimation of the D. Klatt Phoneme Duration Model

Kasparaitis¹,

Beniušė²

2016

Informatica

Self Cite

View full text Add to dashboard Cite

Phoneme duration modelling is one of the stages in prosody modelling for text-to-speech systems. The rule-based phoneme duration model proposed by Klatt (1979) is still quite a popular method. One of the main shortcomings of this method is that the values of the parameters are selected in an experimental way. This work proposes a new iterative algorithm for the automatic estimation of the factors for the Klatt model using the corpus of an annotated audio record of the speaker. The phoneme duration models were built for three different Lithuanian speakers. The quality of the estimation of phonemes durations was evaluated by the root mean square error, the mean absolute error and the correlation coefficient.

show abstract

Corpus-Based Hidden Markov Modelling of the Fundamental Frequency of Lithuanian

Vaičiūnas¹,

Raškinis²,

Kazlauskienė³

2016

Informatica

View full text Add to dashboard Cite

This paper presents the corpus-driven approach in building the computational model of fundamental frequency, or F 0 , for Lithuanian language. The model was obtained by training the HMM-based speech synthesis system HTS on six hours of speech coming from multiple speakers. Several gender specific models, using different parameters and different contextual factors, were investigated. The models were evaluated by synthesizing F 0 contours and by comparing them to the original F 0 contours using criteria of root mean square error (RMSE) and voicing classification error. The HMM-based models showed an improvement of the RMSE over the mean-based model that predicted F 0 of the vowel on the basis of its average normalized pitch.

show abstract

Building Text Corpus for Unit Selection Synthesis

Cited by 4 publications

References 10 publications

Phoneme vs. Diphone in Unit Selection TTS of Lithuanian

Phoneme vs. Diphone in Unit Selection TTS of Lithuanian

Automatic Parameters Estimation of the D. Klatt Phoneme Duration Model

Corpus-Based Hidden Markov Modelling of the Fundamental Frequency of Lithuanian

Contact Info

Product

Resources

About