Einar Meister
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Einar Meister has a total of 45 co-authors. Here are the top co-authors by number of shared publications:
- Lya Meister
Tallinna Tehnikaülikool / Tallinn University of Technology
7 shared publications - Jaan Ross
Estonian Academy of Music and Theatre, Tallinn, Estonia
5 shared publications - Allan Vurma
Estonian Academy of Music and Theatre, Tallinn, Estonia
4 shared publications - Marju Raju
Estonian Academy of Music and Theatre, Tallinn, Estonia
4 shared publications Tuuri Dede
Estonian Academy of Music and Theatre, Tallinn, Estonia
4 shared publications- Pärtel Lippus
Tartu Ülikool / University of Tartu
3 shared publications
Latest Publications
- 10 October 2025
- Published by University of California Press in Music Perception: An Interdisciplinary Journal
- Vol. 43 (4) , 403-417
Managing Reverberant Acoustics in Singing by Extending the Plosive Closures in Vowel–Plosive–Vowel Sequences
Poor intelligibility of sung text often occurs in reverberant rooms due to masking by the reverberation tail of the singer’s voice. This study investigates whether elongating the plosive closure phase can improve the recognition of voiceless plosives in vowel–plosive–vowel sequences sung in reverberant rooms. We hypothesize that a longer plosive closure allows the reverberation tail from the preceding vowel to decay before the plosive burst, thus reducing masking and enhancing plosive recognition. In Experiment 1, 34 listeners heard stimuli (sung single-pitch vowel–plosive–vowel sequences) via headphones, with artificial reverberation and/or Brown Noise added to simulate different acoustics. Experiment 2 involved 33 listeners in a concert hall, where stimuli were played from a loudspeaker on the stage, and Brown Noise was played from a separate sound system. The plosive closure phase in the stimuli was edited using PRAAT software to durations 60 ms, 150 ms, or 260 ms. Recognition of plosives improved by up to 25 percentage points with longer closure phases, depending on the acoustic condition, burst intensity, and vowel pitch. Older listeners, and listeners seated in the back rows of the concert hall, showed poorer recognition. Extending the plosive closure phase generally did not improve plosive recognition in non-reverberant acoustics.
- 10 October 2025
- Published by University of California Press in Music Perception: An Interdisciplinary Journal
- Vol. 43 (4) , 403-417
Managing Reverberant Acoustics in Singing by Extending the Plosive Closures in Vowel–Plosive–Vowel Sequences
Poor intelligibility of sung text often occurs in reverberant rooms due to masking by the reverberation tail of the singer’s voice. This study investigates whether elongating the plosive closure phase can improve the recognition of voiceless plosives in vowel–plosive–vowel sequences sung in reverberant rooms. We hypothesize that a longer plosive closure allows the reverberation tail from the preceding vowel to decay before the plosive burst, thus reducing masking and enhancing plosive recognition. In Experiment 1, 34 listeners heard stimuli (sung single-pitch vowel–plosive–vowel sequences) via headphones, with artificial reverberation and/or Brown Noise added to simulate different acoustics. Experiment 2 involved 33 listeners in a concert hall, where stimuli were played from a loudspeaker on the stage, and Brown Noise was played from a separate sound system. The plosive closure phase in the stimuli was edited using PRAAT software to durations 60 ms, 150 ms, or 260 ms. Recognition of plosives improved by up to 25 percentage points with longer closure phases, depending on the acoustic condition, burst intensity, and vowel pitch. Older listeners, and listeners seated in the back rows of the concert hall, showed poorer recognition. Extending the plosive closure phase generally did not improve plosive recognition in non-reverberant acoustics.
- 1 October 2025
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 158 (4) , 3120-3132
The role of voiced consonant duration in sung vowel-consonant and consonant-vowel recognition
Sung text intelligibility is often a problem, especially in reverberant acoustics, at high pitch, and in the presence of a loud accompaniment. This study aims to discover whether elongating the duration of voiced consonants /m/, /n/, /l/, and /v/ in sung vowel-consonant (VC) and consonant-vowel (CV) sequences improves their recognition. Perception tests were conducted with 42 participants, and the data were analyzed using generalized linear mixed models. Results showed that consonant durations of 20–35 ms are sufficient for a near-perfect recognition when singing at close-to-speaking pitch in non-reverberant acoustics and without accompaniment. However, in reverberant acoustics for VC sequences, longer consonant durations allow the reverberation from the preceding vowel to fade more fully during the consonant, reducing masking by the room reverberation and thereby improving recognition. Also, elongating consonant duration up to about 200 ms (or even longer) improved consonant recognition in the case of stimuli with added Brown Noise, imitating orchestral accompaniment, whereas only negligible impact on the order of succession (CV versus VC) was observed. Recognition tended to be poorer at higher pitch and with longer reverberation.
- 1 October 2025
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 158 (4) , 3120-3132
The role of voiced consonant duration in sung vowel-consonant and consonant-vowel recognition
Sung text intelligibility is often a problem, especially in reverberant acoustics, at high pitch, and in the presence of a loud accompaniment. This study aims to discover whether elongating the duration of voiced consonants /m/, /n/, /l/, and /v/ in sung vowel-consonant (VC) and consonant-vowel (CV) sequences improves their recognition. Perception tests were conducted with 42 participants, and the data were analyzed using generalized linear mixed models. Results showed that consonant durations of 20–35 ms are sufficient for a near-perfect recognition when singing at close-to-speaking pitch in non-reverberant acoustics and without accompaniment. However, in reverberant acoustics for VC sequences, longer consonant durations allow the reverberation from the preceding vowel to fade more fully during the consonant, reducing masking by the room reverberation and thereby improving recognition. Also, elongating consonant duration up to about 200 ms (or even longer) improved consonant recognition in the case of stimuli with added Brown Noise, imitating orchestral accompaniment, whereas only negligible impact on the order of succession (CV versus VC) was observed. Recognition tended to be poorer at higher pitch and with longer reverberation.
- 1 December 2023
- Published by Estonian Literary Museum Scholarly Press in Mäetagused
- Vol. 87, 31-70
Vokaalide ja klusiilide intensiivsussuhted laulmisel teksti arusaadavuse mõjutajana
- 1 December 2023
- Published by Estonian Literary Museum Scholarly Press in Mäetagused
- Vol. 87, 31-70
Vokaalide ja klusiilide intensiivsussuhted laulmisel teksti arusaadavuse mõjutajana
- 1 October 2023
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 154 (4) , 2653-2664
The intensities of vowels and plosive bursts and their impact on text intelligibility in singing
In classical singing, there are often problems with the intelligibility of sung text. The present study aims to test the hypotheses that (1) in loud operatic singing, compared with speaking, the intensity of voiceless plosives increases less than the intensity of vowels, leading to poorer recognition of plosives; and (2) pronouncing the plosive bursts with greater intensity improves their recognition. The acoustic analysis of nine opera arias in Italian from the Classical and Romantic periods performed by ten classically trained singers showed that the average difference in the intensity of vowels when sung and spoken was 14.6 dB [standard deviation (SD) = 7.2 dB], while the difference in the intensity of voiceless plosive bursts was only 6.6 dB (SD = 6 dB). In a perception test with 73 participants, increasing the intensity of the plosive bursts generally improved the recognition of plosives in the sung /a–plosive–a/ sequences, but mainly when reverberation and/or pink noise imitating instrumental accompaniments were added to the stimuli. At the same time, recognition of plosives was often better than chance even when the plosive burst was missing and replaced by silence.
- 1 October 2023
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 154 (4) , 2653-2664
The intensities of vowels and plosive bursts and their impact on text intelligibility in singing
In classical singing, there are often problems with the intelligibility of sung text. The present study aims to test the hypotheses that (1) in loud operatic singing, compared with speaking, the intensity of voiceless plosives increases less than the intensity of vowels, leading to poorer recognition of plosives; and (2) pronouncing the plosive bursts with greater intensity improves their recognition. The acoustic analysis of nine opera arias in Italian from the Classical and Romantic periods performed by ten classically trained singers showed that the average difference in the intensity of vowels when sung and spoken was 14.6 dB [standard deviation (SD) = 7.2 dB], while the difference in the intensity of voiceless plosive bursts was only 6.6 dB (SD = 6 dB). In a perception test with 73 participants, increasing the intensity of the plosive bursts generally improved the recognition of plosives in the sung /a–plosive–a/ sequences, but mainly when reverberation and/or pink noise imitating instrumental accompaniments were added to the stimuli. At the same time, recognition of plosives was often better than chance even when the plosive burst was missing and replaced by silence.
- 31 May 2022
- Published by University of Tartu in Eesti ja soome-ugri keeleteaduse ajakiri. Journal of Estonian and Finno-Ugric Linguistics
- Vol. 13 (2) , 79–118
Articulatory properties of Estonian palatalization by Russian L1 speakers
We studied the articulatory and temporal properties of Estonian palatalization by Russian L1 Estonian L2 speakers and compared the results to Estonian native speakers. The tongue movement of 24 Russian L1 and 21 Estonian L1 speakers were recorded with an electromagnetic articulograph. The speakers articulated phonologically contrastive word pairs and i-stemmed nouns where the final consonant was palatalized. Previous research has shown that palatalization is generally realized by raising the tongue dorsum towards the hard palate. Consequently, the tongue becomes more anterior and wider. This rising of the tongue also lengthens the duration of the segments. The results showed that Russian L1 speakers’ tongue dorsum was higher and more anterior when palatalizing consonants and the subsequent vowels. Compared to native speakers, their tongue was retracted and lower. Russian L1 speakers’ duration of palatalized consonants and the vowels that preceded them were significantly shorter than in the control group. They did not use duration to differentiate palatalization in word pairs as native speakers do. Kokkuvõte. Anton Malmi, Pärtel Lippus, Einar Meister: Vene emakeelega kõnelejate Eesti palatalisatsiooni artikulatoorsed ja temporaalsed tunnused. Artiklis uuritakse, kuidas vene emakeelega kõnelejad hääldavad eesti keele palatalisatsiooni. Varasemad uurimused on näidanud, et konsonanti palataliseeritakse keeleselja tõstmisega kõva suulae poole. Sellega kaasneb ka eespoolsem keele asend ja laiem keele jaotus suulaes ning hääliku pikem kestus. Uurimismaterjal koosnes sõnadest, kus palatalisatsioon eristas tähendust ja i-tüvelistest nimisõnadest, mis lõppesid pika palataliseeritud konsonandiga. Testsõnu sisaldavaid lauseid salvestasime elektromagnetartikulograafiga. Katses osales 24 vene emakeelega ja 21 eesti emakeelega kõnelejat. Tulemused näitasid, et vene emakeelega kõnelejate keel oli palataliseerimisele omaselt kõrgemal ja eespoolsem. Eestlastega võrreldes oli nende keel aga madalamal ja tagapoolsem. Vene emakeelega kõnelejad ei kasutanud palatalisatsioonipaarides kestust eristava tunnusena nagu emakeelsed kõnelejad seda teevad. Nende vokaalid ja konsonandid olid palatalsatsioonipaarides ja i-tüvelistes nimisõnades lühemad kui emakeelsetel kõnelejatel.
- 31 May 2022
- Published by University of Tartu in Eesti ja soome-ugri keeleteaduse ajakiri. Journal of Estonian and Finno-Ugric Linguistics
- Vol. 13 (2) , 79–118
Articulatory properties of Estonian palatalization by Russian L1 speakers
We studied the articulatory and temporal properties of Estonian palatalization by Russian L1 Estonian L2 speakers and compared the results to Estonian native speakers. The tongue movement of 24 Russian L1 and 21 Estonian L1 speakers were recorded with an electromagnetic articulograph. The speakers articulated phonologically contrastive word pairs and i-stemmed nouns where the final consonant was palatalized. Previous research has shown that palatalization is generally realized by raising the tongue dorsum towards the hard palate. Consequently, the tongue becomes more anterior and wider. This rising of the tongue also lengthens the duration of the segments. The results showed that Russian L1 speakers’ tongue dorsum was higher and more anterior when palatalizing consonants and the subsequent vowels. Compared to native speakers, their tongue was retracted and lower. Russian L1 speakers’ duration of palatalized consonants and the vowels that preceded them were significantly shorter than in the control group. They did not use duration to differentiate palatalization in word pairs as native speakers do. Kokkuvõte. Anton Malmi, Pärtel Lippus, Einar Meister: Vene emakeelega kõnelejate Eesti palatalisatsiooni artikulatoorsed ja temporaalsed tunnused. Artiklis uuritakse, kuidas vene emakeelega kõnelejad hääldavad eesti keele palatalisatsiooni. Varasemad uurimused on näidanud, et konsonanti palataliseeritakse keeleselja tõstmisega kõva suulae poole. Sellega kaasneb ka eespoolsem keele asend ja laiem keele jaotus suulaes ning hääliku pikem kestus. Uurimismaterjal koosnes sõnadest, kus palatalisatsioon eristas tähendust ja i-tüvelistest nimisõnadest, mis lõppesid pika palataliseeritud konsonandiga. Testsõnu sisaldavaid lauseid salvestasime elektromagnetartikulograafiga. Katses osales 24 vene emakeelega ja 21 eesti emakeelega kõnelejat. Tulemused näitasid, et vene emakeelega kõnelejate keel oli palataliseerimisele omaselt kõrgemal ja eespoolsem. Eestlastega võrreldes oli nende keel aga madalamal ja tagapoolsem. Vene emakeelega kõnelejad ei kasutanud palatalisatsioonipaarides kestust eristava tunnusena nagu emakeelsed kõnelejad seda teevad. Nende vokaalid ja konsonandid olid palatalsatsioonipaarides ja i-tüvelistes nimisõnades lühemad kui emakeelsetel kõnelejatel.
- Open Access
- 13 May 2022
- Published by Cambridge University Press (CUP) in Journal of the International Phonetic Association
- Vol. 53 (3) , 748-773
Spectral and temporal properties of Estonian palatalization
The aim of this paper is to describe and compare the spectral and temporal properties of Estonian palatalized and non-palatalized consonants /l n s t/ and the vowels that precede them. Acoustic recordings of 43 native Estonian subjects producing word pairs where palatalization differentiated meaning were analyzed in this study. We offer a new perspective on how palatalization is realized by employing a dynamic analysis of the formant trajectories in order to objectively quantify how far the scope of palatalization reaches in the preceding vowel. Results showed that, as in other languages, the most persisting correlate of Estonian palatalization is the rise in F2 in the preceding vowel, where the values are almost always higher already from the beginning. F2 values are higher for /l/ and lower for /n/. The COG (center of gravity) of /s/ is lower in the beginning but rises in the middle. The COG of the burst of /t/ is lower in the onset. As expected, the duration of the vowels preceding consonants is longer, presumably because of the palatalization gesture, but the duration of the consonants does not show a systematic pattern with palatalization.
- 13 May 2022
- Published by Cambridge University Press (CUP) in Journal of the International Phonetic Association
- Vol. 53 (3) , 748-773
Open AccessSpectral and temporal properties of Estonian palatalization
The aim of this paper is to describe and compare the spectral and temporal properties of Estonian palatalized and non-palatalized consonants /l n s t/ and the vowels that precede them. Acoustic recordings of 43 native Estonian subjects producing word pairs where palatalization differentiated meaning were analyzed in this study. We offer a new perspective on how palatalization is realized by employing a dynamic analysis of the formant trajectories in order to objectively quantify how far the scope of palatalization reaches in the preceding vowel. Results showed that, as in other languages, the most persisting correlate of Estonian palatalization is the rise in F2 in the preceding vowel, where the values are almost always higher already from the beginning. F2 values are higher for /l/ and lower for /n/. The COG (center of gravity) of /s/ is lower in the beginning but rises in the middle. The COG of the burst of /t/ is lower in the onset. As expected, the duration of the vowels preceding consonants is longer, presumably because of the palatalization gesture, but the duration of the consonants does not show a systematic pattern with palatalization.
- Open Access
- 10 March 2021
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 9 (3) , e21331
Detection of Bulbar Involvement in Patients With Amyotrophic Lateral Sclerosis by Machine Learning Voice Analysis: Diagnostic Decision Support Development Study
Background: Bulbar involvement is a term used in amyotrophic lateral sclerosis (ALS) that refers to motor neuron impairment in the corticobulbar area of the brainstem, which produces a dysfunction of speech and swallowing. One of the earliest symptoms of bulbar involvement is voice deterioration characterized by grossly defective articulation; extremely slow, laborious speech; marked hypernasality; and severe harshness. Bulbar involvement requires well-timed and carefully coordinated interventions. Therefore, early detection is crucial to improving the quality of life and lengthening the life expectancy of patients with ALS who present with this dysfunction. Recent research efforts have focused on voice analysis to capture bulbar involvement.Objective: The main objective of this paper was (1) to design a methodology for diagnosing bulbar involvement efficiently through the acoustic parameters of uttered vowels in Spanish, and (2) to demonstrate that the performance of the automated diagnosis of bulbar involvement is superior to human diagnosis.Methods: The study focused on the extraction of features from the phonatory subsystem—jitter, shimmer, harmonics-to-noise ratio, and pitch—from the utterance of the five Spanish vowels. Then, we used various supervised classification algorithms, preceded by principal component analysis of the features obtained.Results: To date, support vector machines have performed better (accuracy 95.8%) than the models analyzed in the related work. We also show how the model can improve human diagnosis, which can often misdiagnose bulbar involvement.Conclusions: The results obtained are very encouraging and demonstrate the efficiency and applicability of the automated model presented in this paper. It may be an appropriate tool to help in the diagnosis of ALS by multidisciplinary clinical teams, in particular to improve the diagnosis of bulbar involvement.
- 10 March 2021
- Published by JMIR Publications Inc. in JMIR Medical Informatics
- Vol. 9 (3) , e21331
Open AccessDetection of Bulbar Involvement in Patients With Amyotrophic Lateral Sclerosis by Machine Learning Voice Analysis: Diagnostic Decision Support Development Study
Background: Bulbar involvement is a term used in amyotrophic lateral sclerosis (ALS) that refers to motor neuron impairment in the corticobulbar area of the brainstem, which produces a dysfunction of speech and swallowing. One of the earliest symptoms of bulbar involvement is voice deterioration characterized by grossly defective articulation; extremely slow, laborious speech; marked hypernasality; and severe harshness. Bulbar involvement requires well-timed and carefully coordinated interventions. Therefore, early detection is crucial to improving the quality of life and lengthening the life expectancy of patients with ALS who present with this dysfunction. Recent research efforts have focused on voice analysis to capture bulbar involvement.Objective: The main objective of this paper was (1) to design a methodology for diagnosing bulbar involvement efficiently through the acoustic parameters of uttered vowels in Spanish, and (2) to demonstrate that the performance of the automated diagnosis of bulbar involvement is superior to human diagnosis.Methods: The study focused on the extraction of features from the phonatory subsystem—jitter, shimmer, harmonics-to-noise ratio, and pitch—from the utterance of the five Spanish vowels. Then, we used various supervised classification algorithms, preceded by principal component analysis of the features obtained.Results: To date, support vector machines have performed better (accuracy 95.8%) than the models analyzed in the related work. We also show how the model can improve human diagnosis, which can often misdiagnose bulbar involvement.Conclusions: The results obtained are very encouraging and demonstrate the efficiency and applicability of the automated model presented in this paper. It may be an appropriate tool to help in the diagnosis of ALS by multidisciplinary clinical teams, in particular to improve the diagnosis of bulbar involvement.
- Open Access
- 11 September 2020
- Published by JMIR Publications Inc. in JMIR Preprints
Detection of Bulbar Involvement in Patients With Amyotrophic Lateral Sclerosis by Machine Learning Voice Analysis: Diagnostic Decision Support Development Study (Preprint)
BACKGROUND Bulbar involvement is a term used in amyotrophic lateral sclerosis (ALS) that refers to motor neuron impairment in the corticobulbar area of the brainstem, which produces a dysfunction of speech and swallowing. One of the earliest symptoms of bulbar involvement is voice deterioration characterized by grossly defective articulation; extremely slow, laborious speech; marked hypernasality; and severe harshness. Bulbar involvement requires well-timed and carefully coordinated interventions. Therefore, early detection is crucial to improving the quality of life and lengthening the life expectancy of patients with ALS who present with this dysfunction. Recent research efforts have focused on voice analysis to capture bulbar involvement. OBJECTIVE The main objective of this paper was (1) to design a methodology for diagnosing bulbar involvement efficiently through the acoustic parameters of uttered vowels in Spanish, and (2) to demonstrate that the performance of the automated diagnosis of bulbar involvement is superior to human diagnosis. METHODS The study focused on the extraction of features from the phonatory subsystem—jitter, shimmer, harmonics-to-noise ratio, and pitch—from the utterance of the five Spanish vowels. Then, we used various supervised classification algorithms, preceded by principal component analysis of the features obtained. RESULTS To date, support vector machines have performed better (accuracy 95.8%) than the models analyzed in the related work. We also show how the model can improve human diagnosis, which can often misdiagnose bulbar involvement. CONCLUSIONS The results obtained are very encouraging and demonstrate the efficiency and applicability of the automated model presented in this paper. It may be an appropriate tool to help in the diagnosis of ALS by multidisciplinary clinical teams, in particular to improve the diagnosis of bulbar involvement.
- 11 September 2020
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDetection of Bulbar Involvement in Patients With Amyotrophic Lateral Sclerosis by Machine Learning Voice Analysis: Diagnostic Decision Support Development Study (Preprint)
BACKGROUND Bulbar involvement is a term used in amyotrophic lateral sclerosis (ALS) that refers to motor neuron impairment in the corticobulbar area of the brainstem, which produces a dysfunction of speech and swallowing. One of the earliest symptoms of bulbar involvement is voice deterioration characterized by grossly defective articulation; extremely slow, laborious speech; marked hypernasality; and severe harshness. Bulbar involvement requires well-timed and carefully coordinated interventions. Therefore, early detection is crucial to improving the quality of life and lengthening the life expectancy of patients with ALS who present with this dysfunction. Recent research efforts have focused on voice analysis to capture bulbar involvement. OBJECTIVE The main objective of this paper was (1) to design a methodology for diagnosing bulbar involvement efficiently through the acoustic parameters of uttered vowels in Spanish, and (2) to demonstrate that the performance of the automated diagnosis of bulbar involvement is superior to human diagnosis. METHODS The study focused on the extraction of features from the phonatory subsystem—jitter, shimmer, harmonics-to-noise ratio, and pitch—from the utterance of the five Spanish vowels. Then, we used various supervised classification algorithms, preceded by principal component analysis of the features obtained. RESULTS To date, support vector machines have performed better (accuracy 95.8%) than the models analyzed in the related work. We also show how the model can improve human diagnosis, which can often misdiagnose bulbar involvement. CONCLUSIONS The results obtained are very encouraging and demonstrate the efficiency and applicability of the automated model presented in this paper. It may be an appropriate tool to help in the diagnosis of ALS by multidisciplinary clinical teams, in particular to improve the diagnosis of bulbar involvement.
- Open Access
- 31 December 2015
- Published by University of Tartu in Eesti ja soome-ugri keeleteaduse ajakiri. Journal of Estonian and Finno-Ugric Linguistics
- Vol. 6 (3) , 79-96
Production of Estonian quantity contrasts by Japanese speakers
Estonian and Japanese are quantity languages both exploiting the duration cue to implement phonological contrasts. However, the quantity systems of the two languages are different – Estonian features a three-way quantity contrast while Japanese has a binary contrast. This paper studies how L2 subjects with Japanese-language background (L2-JP) produce the Estonian quantity contrasts. For the acoustic analysis the speech recordings by six L2-JP subjects and 12 native Estonian (L1-EE) subjects were used. The material analyzed consists of read sentences comprising triplets of segmentally identical disyllabic target words in the quantities Q1 (short), Q2 (long) and Q3 (overlong). In their production, the L2-JP subjects successfully produced the Q1/Q2 contrast but failed in contrasting vocalic Q2 and Q3 (CVVCV vs. CVVːCV) oppositions; however, the subjects managed to produce the Q2/Q3 consonantal quantity contrasts (CVCCV vs. CVCːCV). The L2-JP subjects’ segment durations differing from those of the L1-EE subjects, reveal the role of native durational patterns on the acquisition of Estonian quantity oppositions.Kokkuvõte. Einar Meister, Rena Nemoto, Lya Meister: Eesti välted jaapani emakeelega keelejuhtide häälduses. Eesti ja jaapani keel on kvantiteedikeeled, st mõlemad keeled kasutavad kestust fonoloogiliste vastanduste väljendamiseks. Kvantiteedisüsteemid kahes keeles on siiski erinevad – eesti keelt iseloomustab kolmene vältevastandus, jaapani keeles on kestusvastandus binaarne. Artiklis uuriti eesti väldete hääldust jaapani emakeelega keelejuhtide kõnes ja võrreldi seda eesti emakeelega keelejuhtide hääldusega. Akustiliseks analüüsiks kasutati kuue eesti keelt võõrkeelena kõneleva jaapanlase ja 12 eesti emakeelega keelejuhi etteloetud lausete salvestusi, milles esinesid vältevastandusi kandvad kahesilbilised sõnad. Tulemused näitasid, et jaapani keelejuhid hääldasid kontrastiivselt esma- ja teisevältelisi sõnu, kuid ei eristanud teise- ja kolmandavältelisi vokaalikeskse vastandusega (CVVCV vs CVVːCV) sõnu. Samas hääldasid nad erinevalt teise- ja kolmandavältelisi konsonandikeskse vastandusega (CVCCV vs CVCːCV) sõnu. Leitud segmendikestuste erinevused jaapanlaste ja eestlaste kõnes peegeldavad emakeelsete kestusmustrite erinevusi kahes keeles.Märksõnad: eesti keel, jaapani keel, aktsendiga kõne, fonoloogiline kestus, välted
- 31 December 2015
- Published by University of Tartu in Eesti ja soome-ugri keeleteaduse ajakiri. Journal of Estonian and Finno-Ugric Linguistics
- Vol. 6 (3) , 79-96
Open AccessProduction of Estonian quantity contrasts by Japanese speakers
Estonian and Japanese are quantity languages both exploiting the duration cue to implement phonological contrasts. However, the quantity systems of the two languages are different – Estonian features a three-way quantity contrast while Japanese has a binary contrast. This paper studies how L2 subjects with Japanese-language background (L2-JP) produce the Estonian quantity contrasts. For the acoustic analysis the speech recordings by six L2-JP subjects and 12 native Estonian (L1-EE) subjects were used. The material analyzed consists of read sentences comprising triplets of segmentally identical disyllabic target words in the quantities Q1 (short), Q2 (long) and Q3 (overlong). In their production, the L2-JP subjects successfully produced the Q1/Q2 contrast but failed in contrasting vocalic Q2 and Q3 (CVVCV vs. CVVːCV) oppositions; however, the subjects managed to produce the Q2/Q3 consonantal quantity contrasts (CVCCV vs. CVCːCV). The L2-JP subjects’ segment durations differing from those of the L1-EE subjects, reveal the role of native durational patterns on the acquisition of Estonian quantity oppositions.Kokkuvõte. Einar Meister, Rena Nemoto, Lya Meister: Eesti välted jaapani emakeelega keelejuhtide häälduses. Eesti ja jaapani keel on kvantiteedikeeled, st mõlemad keeled kasutavad kestust fonoloogiliste vastanduste väljendamiseks. Kvantiteedisüsteemid kahes keeles on siiski erinevad – eesti keelt iseloomustab kolmene vältevastandus, jaapani keeles on kestusvastandus binaarne. Artiklis uuriti eesti väldete hääldust jaapani emakeelega keelejuhtide kõnes ja võrreldi seda eesti emakeelega keelejuhtide hääldusega. Akustiliseks analüüsiks kasutati kuue eesti keelt võõrkeelena kõneleva jaapanlase ja 12 eesti emakeelega keelejuhi etteloetud lausete salvestusi, milles esinesid vältevastandusi kandvad kahesilbilised sõnad. Tulemused näitasid, et jaapani keelejuhid hääldasid kontrastiivselt esma- ja teisevältelisi sõnu, kuid ei eristanud teise- ja kolmandavältelisi vokaalikeskse vastandusega (CVVCV vs CVVːCV) sõnu. Samas hääldasid nad erinevalt teise- ja kolmandavältelisi konsonandikeskse vastandusega (CVCCV vs CVCːCV) sõnu. Leitud segmendikestuste erinevused jaapanlaste ja eestlaste kõnes peegeldavad emakeelsete kestusmustrite erinevusi kahes keeles.Märksõnad: eesti keel, jaapani keel, aktsendiga kõne, fonoloogiline kestus, välted
- 1 October 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 141-143
Real-time mimicking of estonian speaker's mouth movements on a 3D avatar using Kinect 2
In this paper an algorithm for the real-time imitation of lips movements of an Estonian speaker on a 3D avatar is proposed. This research is focused on obtaining the RGB and depth information from a Kinect 2 camera, processing it and moving the avatar accordingly. For the more stable extraction of information about lip shape and movement, cosine function is used. The depth info helps to determine the closest speaker and stabilize the movement of avatar and tracking of lips.
- 1 October 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 141-143
Real-time mimicking of estonian speaker's mouth movements on a 3D avatar using Kinect 2
In this paper an algorithm for the real-time imitation of lips movements of an Estonian speaker on a 3D avatar is proposed. This research is focused on obtaining the RGB and depth information from a Kinect 2 camera, processing it and moving the avatar accordingly. For the more stable extraction of information about lip shape and movement, cosine function is used. The depth info helps to determine the closest speaker and stabilize the movement of avatar and tracking of lips.