Lithuanian Speech Recognition Using the English Recognizer

Kasparaitis, Pijus

doi:10.15388/informatica.2008.227

Cited by 17 publications

(6 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(4), (5) where Q ' , Q'', S ' ,…Y'' are constant values. The forces from kinematics stimulations q(t), ψ(t) and y(t) include into the Lagrange equation ( 1) by potential energy Π and dissipative function Φ.…”

Section: Modeling Of Measuring Processesmentioning

confidence: 99%

Dynamic Errors of CMM Probes

Kasparaitis

Šukys

2006

SSP

View full text Add to dashboard Cite

Here are proposed methods to evaluate the accuracy of measuring heads (MH) of coordinate measuring machines (CMM). Kinematic touch-trigger, piezoresonant touch-trigger and scanning 3D probes are investigated as the most often being used for the 3D measurements. The dominating errors of kinematic touch-trigger probes are expressed by the polygonal (triangular) characteristics of their kinematics and by random dispersion of the contact forces. These features characterize the piezoresonant touch-trigger probes as well but instability of resonance of the system components are also presented here. Dominating error’s characteristics of the scanning probes consist of the instability of pushing the probe stylus out of the measured surface by dynamical forces. Some mathematical models, space error’s approximation and results of experimental investigation of the measuring heads are presented. It may be used for the improvement of CMM and other measuring equipment.

show abstract

Section: Modeling Of Measuring Processesmentioning

confidence: 99%

Dynamic Errors of CMM Probes

Kasparaitis

Šukys

2006

SSP

View full text Add to dashboard Cite

show abstract

“…Prior to that study, the phonology of Polish was described in many sources (e.g., Gussmann, 2007;Jassem, 2003;Oliver, Szklanny, 2006). It should also be noted that much effort was performed by several Polish and Lithuanian research centers aiming at speech recognition, a few examples of which are given in here: (Kłosowski et al, 2014), analysis of acoustics speech properties (Izydorczyk, Kłosowski, 2001), adaptation of foreign language speech recognition engines for Lithuanian speech recognition (Rudzionis et al, 2009;Kasparaitis, 2008), development of phonemic language corpus for Polish (Kłosowski, 2017) by employing automatic grapheme-to-phoneme conversion of the source orthographic language corpus, obtained from the National Corpus of Polish (NCP) (Przepiórkowski et al, 2012), creating Polish phoneme statistics (Ziółko et al, 2009;2014), etc.…”

Section: Introductionmentioning

confidence: 99%

Archives of Acoustics

Korvel

Kurasova

Kostek

2019

View full text Add to dashboard Cite

The goal of this research is to find a set of acoustic parameters that are related to differences between Polish and Lithuanian language consonants. In order to identify these differences, an acoustic analysis is performed, and the phoneme sounds are described as the vectors of acoustic parameters. Parameters known from the speech domain as well as those from the music information retrieval area are employed. These parameters are time-and frequency-domain descriptors. English language as an auxiliary language is used in the experiments. In the first part of the experiments, an analysis of Lithuanian and Polish language samples is carried out, features are extracted, and the most discriminating ones are determined. In the second part of the experiments, automatic classification of Lithuanian/English, Polish/English, and Lithuanian/Polish phonemes is performed.

show abstract

“…Using feature vectors, the back-end of ASR system selects the most consistent word from the reference database. Plenty of ASR research studies have investigated back-end classifiers (Kasparaitis, 2008;Lipeika et al, 2003;Norkin and Keyzer, 2009;Bartkute, 2009). Contrary, we concentrate on feature quality estimation in the front-end of ASR system.…”

Section: Introductionmentioning

confidence: 99%

Metrics Based Quality Estimation of Speech Recognition Features

Lileikytė¹,

Telksnys²

2013

Informatica

View full text Add to dashboard Cite

The performance of an automatic speech recognition system heavily depends on the used feature set. Quality of speech recognition features is estimated by classification error, but then the recognition experiments must be performed, including both front-end and back-end implementations. We propose a method for features quality estimation that does not require recognition experiments and accelerate automatic speech recognition system development. The key component of our method is usage of metrics right after front-end features computation. The experimental results show that our method is suitable for recognition systems with back-end Euclidean space classifiers.

show abstract

Lithuanian Speech Recognition Using the English Recognizer

Cited by 17 publications

References 9 publications

Dynamic Errors of CMM Probes

Dynamic Errors of CMM Probes

Archives of Acoustics

Metrics Based Quality Estimation of Speech Recognition Features

Contact Info

Product

Resources

About