Optimization of Lithuanian Diphone Databases
Journal Title: Kalbu studijos / Studies about Languages - Year 2011, Vol 19, Issue 0
Abstract
Creation and optimization of the Lithuanian diphone inventory used for concatenative text-to-speech synthesis is studied in this paper. Creation of diphone database starts with compilation of a list of diphones. This is not a trivial problem, because some diphones are not valid. If valid diphones are deduced using language phonotactics rules, some diphones which needed to synthesize foreign words are omitted. Besides, a lot of practically unused diphones are added to the list. Statistical diphone usage analysis was performed in this work. The results of this analysis imply that using statistically motivated diphone inventory pruning one can compile a much smaller inventory while keeping a very high text coverage. Diphone inventory pruning using phone similarity (one diphone is substituted with another that sounds similarly) and using phone stretching (a missing diphone is synthesized by stretching phones of adjacent diphones) is described, as well. Listening experiments with diphones that contain a vowel or diphthong followed by stop consonant (or fricative, for phone stretching) were executed. Groups of diphones were identified for which synthesized speech quality is not degraded (and for which quality is degraded only marginally) when using two described methods. In addition, potential diphone inventory reduction was estimated.
Authors and Affiliations
Pijus Kasparaitis, Žilvinas Ledas
Keywords
Related Articles
- EP ID EP145299
- DOI -
- Views 93
- Downloads 0
Relevance Theory and Pragmatics of Discourse Connectives
Connectives are very frequent words. According to the data of the Current Lithuanian Language Corpus, the most of them fall within one hundred of the most common words in the frequency word list of the Lithuanian languag...
A Cognitive Approach to Teaching English for Special Purposes (ESP)
Cognitive science today makes it possible to combine independent studies under the auspices of one focal problem — the workings of the human mind in knowledge acquisition. Foreign language teaching draws heavily on ling...
Authentic Resources in Technology-based ESP Learning
The authors of the article aim to present ESP students’ attitude towards and impact of different authentic materials on their motivation, to identify advantages and disadvantages of using these authentic resources for l...
Lietuvių kultūrinių teksto reikšmių interpretacija ir vertimas
Straipsnyje nagrinėjamos eksplicitiškai ir implicitiškai perteiktos lietuvių literatūros kūrinių kultūrinės reikšmės, jų interpretacija ir vertimas į prancūzų kalbą. Taikant gretinamąjį, kiekybinės analizės ir aprašomąj...
Naujausių kompiuterijos ir interneto terminų apžvalga
Informacinių technologijų (IT) sritis yra ypatinga tuo, kad nauji terminai joje atsiranda kiekvieną dieną. Yra paskaičiuota, kad anglų kalboje per dieną atsiranda apie 15 naujų žodžių, iš kurių 5 yra susiję su IT. Ši&nbs...