Using bands of frequencies for vowel recognition for Polish language

Płonkowski, M.

doi:10.1007/s10772-014-9259-z

Cited by 6 publications

(2 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our future work we will analyze this normalization procedure regarding the Polish language in which some elements are different from English [16,17]. Hence they require separate analysis.…”

Section: Discussionmentioning

confidence: 99%

The use of pitch in Large-Vocabulary Continuous Speech Recognition System

Płonkowski

Урбанович

2016

ELECTROTECHNICAL REVIEW

Self Cite

View full text Add to dashboard Cite

In this article the authors normalize the speech signal based on the publicly available AN4 database. The authors added to the algorithm of calculating the MFCC coefficients, the normalization procedure, that uses pitch of the voice. As demonstrated by empirical tests authors were able to improve speech recognition accuracy rate of about 20%.Streszczenie. W niniejszym artykule autorzy normalizują sygnał mowy wykorzystując publicznie dostępną bazę danych AN4. Autorzy dodali do algorytmu obliczania współczynników MFCC, procedurę normalizacji, wykorzystującą wysokość tonu głosu. Jak wynika z przeprowadzonych testów, autorzy uzyskali poprawę dokładności rozpoznawania mowy o około 20% (Wykorzystanie wysokości tonu głosu w systemach rozpoznawania mowy ciągłej z dużą ilością słów).

show abstract

Section: Discussionmentioning

confidence: 99%

The use of pitch in Large-Vocabulary Continuous Speech Recognition System

Płonkowski

Урбанович

2016

ELECTROTECHNICAL REVIEW

Self Cite

View full text Add to dashboard Cite

show abstract

“…The study used four classifiers, the SpD method produces a vowel recognition accuracy rate of 92.42% to 95.11%. Plonkowski [4] presents a simple method for the introduction of Polish vowels. Martin determines three characteristics of the frequency field, namely average, standard deviation, and the maximum value to recognize each vowel.…”

Section: Introduction To Vowelsmentioning

confidence: 99%

Analysis of the Indonesian Vowel /e/ For Lip Synchronization Animation

Rachman¹,

Hidayat²,

Nugroho³

2017

EECSI

View full text Add to dashboard Cite

Currently, voice recognition technology is widely used to produce lip sync animation. Vowels take the most dominant roles for lip sync animation as it always exists in every syllable. Therefore, it is necessary to select appropriate vowel traits for the system to be accurate. In general, there are five vowels of Indonesian language, namely /a/ /i/ /u/ /e/ and /o/. However, there are two vowels that contain several different tones: /o/ that are pronounced /o/ and /O/, and /e/ that are pronounced /e/, /ǝ/, and /ɛ/. The difference in tone can affect the accuracy of voice recognition on the lip sync animation system if it is not specified further. In this paper, the characteristic values of vowel /e/, /ǝ/, and /ɛ/ are compared and analyzed to find the significance of the difference. The sought characteristic values are the frequency of the formant (F1, F2, and F3) through the Praat software used to extract the features. Comparison is done using a statistical test of t-test. The results show that the three vowel tones /e/ have significant differences for all of F1 and most of F2.

show abstract