Generating segmental foreign accent

For most of us, speaking in a non-native language involves deviating to some extent from native pronunciation norms. However, the detailed basis for foreign accent (FA) remains elusive, in part due to methodological challenges in isolating segmental from suprasegmental factors. The current study examines the role of segmental features in conveying FA through the use of a generative approach in which accent is localised to single consonantal segments. Three techniques are evaluated: the first requires a highly-proficiency bilingual to produce words with isolated accented segments; the second uses cross-splicing of context-dependent consonants from the non-native language into native words; the third employs hidden Markov model synthesis to blend voice models for both languages. Using English and Spanish as the native/non-native languages respectively, listener cohorts from both languages identified words and rated their degree of FA. All techniques were capable of generating accented words, but to differing degrees. Naturally-produced speech led to the strongest FA ratings and synthetic speech the weakest, which we interpret as the outcome of over-smoothing. Nevertheless, the flexibility offered by synthesising localised accent encourages further development of the method.

Estudio acústico y electroacústico para la realización de un musical en el teatro Rialto de Madrid

El propósito de este Proyecto Fin de Carrera es el estudio acústico y electroacústico de la realización del musical “Hoy no me puedo levantar” en el Teatro Rialto de Madrid en 2005. En primer lugar, se realiza una breve introducción histórica, citando sus remodelaciones y comentando la situación actual del recinto. Posteriormente, es analizado el equipo de sonido empleado en el espectáculo a partir de cada uno de los distintos controles de sonido: FOH (Front of Hause), monitores y microfonía inalámbrica. De cada uno de ellos se explican sus principales funciones y los sistemas que los conforman. También se describe la utilización de las cabinas insonorizadas. A continuación, se detallan los sistemas electroacústicos (empleados en el diseño) de la sonorización de dicho musical, que se consideran divididos en las siguientes partes: sistema principal, refuerzos y retardos, efectos y monitores. Además, se detalla el software RMS (Remote Monitoring System), que aporta información del funcionamiento de estos sistemas en tiempo real. Seguidamente, se exponen el equipo, procedimiento y resultados de la medida in situ en el Teatro, aplicando la Norma UNE-EN ISO 3382-2/2008 para obtener el tiempo de reverberación y ruido de fondo. Con el objeto de inicializar la simulación por ordenador, primero se exportan los planos originales de AutoCAD a EASE 4.4, donde se finaliza el proceso de modelar el recinto. Posteriormente, se asignan materiales, áreas de audiencia, puntos de escucha y se ubican los sistemas electroacústicos. Se afina el tiempo de reverberación obtenido en la medida in situ mediante materiales de la base de datos del propio software. También se ajustan los sistemas electroacústicos en el recinto para obtener la ecualización empleada y los niveles de presión sonora directo y total para distintas frecuencias. Una vez finalizados los pasos anteriores, se procede a realizar estudios psicoacústicos para comprobar posibles ecos y el efecto precedencia (empleando retardos electrónicos o delays). Finalmente, se realizan estudios de inteligibilidad, en los que se justifica la Claridad de Voz (C50) y Claridad Musical (C80); el Índice de inteligibilidad del habla (SII), la Pérdida de articulación de consonantes (Alcons) y el Índice de transmisión del habla (STI). Por último se expone el presupuesto del proyecto y del alquiler del equipo de sonido del musical y se exponen las conclusiones del Proyecto Final de Carrera. ABSTRACT. The purpose of this Final Degree Project is the acoustic and electro-acoustic study of the musical “Hoy No Me Puedo Levantar” at Teatro Rialto in 2005 (Madrid, Spain). First of all, a brief review of its history is made, quoting its refurbishments and discussing the current situation of this enclosure. Later, the sound equipment of the show is analyzed through every different sound controls: FOH (Front Of House), monitors and wireless microphones. There is also an explanation about their principal functions and systems, as well as a description of the soundproof cabins. Then, the electro-acoustic systems are detailed and divided in the following parts: main system, boosters and delays, effects and monitors. The RMS software (Remote Monitoring System) is described too, since it gives relevant information of the systems operations in real time. Afterwards, equipment, procedures and results of the measurements are exposed, applying the UNE-EN ISO 3382-2/2008 regulation in order to obtain the reverberation time and background noise of the theatre. With the purpose of initialize the computer simulation, original plans are exported from AutoCad to EASE 4.4., where its modeling process is ended. Materials, audience areas, hearing points and electro-acoustic locations are assigned below. At the same time, reverberation time is tuned up using database materials of the software itself. Also, electro-acoustic systems of the enclosure are adjusted to get the equalization and pressure sound levels of the different frequencies. Once previous steps are finished, psycho-acoustic studies are made to check possible echoes and the precedence effect - using electronic delays -. Finally, intelligibility studies are detailed, where the Voice and Musical Clarities are justified: The Speech Intelligibility Index, the Loss of Consonants Articulation and the Talk Transmission Index. This Final Degree Project ends describing the budget and rent of the sound equipment and the final conclusions.

Influência do uso de amostras de referência no julgamento perceptivo-auditivo da oclusiva glotal

Objetivos: estabelecer amostras de referência constituídas por gravações julgadas com consenso como representativas da presença ou ausência da oclusiva glotal (OG) e comparar julgamentos perceptivo-auditivos da presença e ausência da OG com e sem o uso de amostras de referência. Metodologia: o estudo foi dividido em duas etapas. Durante a ETAPA 1, 480 frases referentes aos sons oclusivos e fricativos produzidas por falantes com história de fissura labiopalatina foram julgadas por três fonoaudiólogas experientes quanto à identificação da OG. As frases foram julgadas individualmente e aquelas que não apresentaram consenso inicial foram julgadas novamente de maneira simultânea. As amostras julgadas com consenso com relação à presença ou ausência da OG durante produção das seis consoantes-alvo oclusivas e seis fricativas foram selecionadas para estabelecer um Banco de Amostras Representativas da OG. A ETAPA 2 consistiu na seleção de 48 amostras de referência referentes aos 12 sons de interesse e 120 amostras experimentais e, o julgamento dessas amostras experimentais por três grupos de juízes, cada grupo com três juízes com experiências distintas com relação ao julgamento de fala na fissura de palato. Os juízes julgaram as amostras experimentais duas vezes, primeiro sem acesso às referências e, após uma semana, com acesso às referências. Resultados: os julgamentos realizados na ETAPA 1 evidenciaram consenso com relação a OG em 352 amostras, sendo 120 frases com produção adequada para os sons de interesse e 232 representativas do uso da OG. Essas 352 amostras constituíram o Banco de amostras Representativas da OG. Os resultados da ETAPA 2 indicaram que ao comparar a média do valor de Kappa obtida para os 12 sons de interesse em cada um dos grupos nos julgamentos sem e com acesso às amostras de referência a concordância para o grupo 1 (G1) passou de regular (K=0,35) para moderada (K=0,55), para o grupo 2 (G2) passou de moderada (K=0,44) para substancial (K=0,76) e para o grupo 3 (G3) passou de substancial (K=0,72) para quase perfeita (K=0,83). Observou-se que as melhores concordâncias ocorreram para o grupo dos fonoaudiólogos experientes (G3), seguido dos fonoaudiólogos recém-formados (G2), com as piores observadas para o grupo de alunos de graduação (G1). Conclusão: um Banco de Amostras de Referência Representativas da OG foi estabelecido e os julgamentos perceptivo-auditivos de juízes com uso das amostras de referência foram obtidos com concordância inter-juízes e porcentagem de acertos melhor do que os julgamentos sem acesso às referências. Os resultados sugerem a importância do uso de amostras de referência para minimizar a subjetividade da avaliação perceptivo auditiva da fala.

El tratado de Teodomiro en su contexto histórico y paleográfico

Tras una revisión de nuestros conocimientos sobre la escritura en árabe en el año de la firma del tratado de Teodomiro, llegamos a la conclusión que estaba en desarrollo este proceso de escritura y que aún no se escribían los puntos diacríticos en las consonantes. Al hilo de ello, el objetivo principal ha sido hacer una relectura de los dos manuscritos principales representativos de las líneas de transmisión del tratado: el de al-‘Uḏrī y el de al-Ḍabbī. Finalmente, se propone un stemma de transmisión de sus textos.

Alfabetic-order table ... altered and fitted with three figures /

Two tables mounted on boards and bound.

The language of the Rushworth gloss to the gospel of Matthew and the Mercian dialect.

Pt.II. The vowels of other syllables than stem-syllables ; consonants ; inflection.

English philology pamphlets

A collection of miscellaneous pamphlets.

Phonological features of Yazghulami : A field study

Yazghulami is a South-East Iranian language spoken in the Pamir area of Tajikistan by about 9000 people. This study gives an account of the phonology of the language by describing contrastive segments and their distribution and realizations, as well as describing suprasegmental features such as syllable structure and stress patterns. Field research was carried out in a community of Yazghulami speakers in Dushanbe, the capital of Tajikistan, by recording, transcribing and annotating spoken language. Yazghulami is analyzed as having 8 vowel phonemes of which one pair contrasts in length, and 36 consonant phonemes with a considerable display of palatal, velar and uvular phonemes, of which a set of three labialized plosives and three labialized fricatives is found. The syllable structure of Yazghulami allows for clusters of no more than two consonants in the onset and two in the coda; clusters in both positions do not occur in one and the same syllable. The stress generally falls on the last syllable of a word, although when nouns are inflected with suffixes, the stress instead falls on the last syllable of the stem. With these results, a foundation for further efforts to develop and increase the status of this endangered language is laid.

Perceptual evaluation of motor speech following treatment for childhood cerebellar tumour

The speech characteristics, oromotor function and speech intelligibility of a group of children treated for cerebellar tumour (CT) was investigated perceptually. Assessment of these areas was performed on 11 children treated for CT with dysarthric speech as well as 21 non-neurologically impaired controls matched for age and sex to obtain a comprehensive perceptual profile of their speech and oromotor mechanism. Contributing to the perception of dysarthria were a number of deviant speech dimensions including imprecision of consonants, hoarseness and decreased pitch variation, as well as a reduction in overall speech intelligibility for both sentences and connected speech. Oromotor assessment revealed deficits in lip, tongue and laryngeal function, particularly relating to deficits in timing and coordination of movements. The most salient features of the dysarthria seen in children treated for CT were the mild nature of the speech disorder and clustering of speech deficits in the prosodic, phonatory and articulatory aspects of speech production.

Comparability of perceptual analysis of speech characteristics in Australian and Swedish speakers with multiple sclerosis

The aims of the present study were to compare the perceptual assessments of deviant speech signs (dysarthria) exhibited by Australian and Swedish speakers with multiple sclerosis (MS) and to explore whether judgements of dysarthria differed depending on whether the speakers and the judges spoke the same or different languages. Ten Australian and 10 Swedish individuals with MS (matched as closely as possible for age, gender, progression type and severity of dysarthria) were assessed by 2 Australian and 2 Swedish clinically experienced judges using a protocol including 33 speech parameters. Results show that the following perceptual dimensions were identified by both pairs of judges in both groups of speakers to a just noticeable or moderate degree: imprecise consonants, inappropriate pitch level, reduced general rate, and glottal fry. The reliability (Spearman rank-order correlation) of the consensus ratings from the Australian and the Swedish judges was high, with a mean rho of 85.7 for the Australian speakers and mean rho of 84.3 for the Swedish speakers. The most difficult perceptual parameters to assess (i.e. to agree on) included harshness, level of pitch and loudness, precision of consonants and general stress pattern. The study indicated that perceptual assessments of speech characteristics in individuals with MS are informative and can be achieved with high inter-judge reliability irrespective of the judge's knowledge of the speaker's language. Copyright (C) 2003 S. Karger AG, Basel.

Articulatory function in hypokinetic dysarthria: An electropalatographic examination of two cases

The present study employed electropalatography (EPG) and a nonspeech measure of lingual function to examine, in detail, the articulatory production deficits of two individuals with Parkinson disease (PD) and hypokinetic dysarthria. Participants read 10 repetitions of CV words contained within the carrier phrase I saw a _ today while wearing an EPG artificial palate. Target consonants included the alveolar stop /t/, lateral approximant /l/, and the alveolar fricative /s/ in the /a/ vowel environment. The results of the two participants were compared to an age-matched control group. Examination of the perceptual features of articulatory production, lingual strength, fine force control and endurance, tongue-palate contact patterns, and segment durations were conducted. Results of the study revealed quite different articulatory deficits in the two participants. Specifically, the articulation of Participant One (P1) was characterized by a fast rate of speech, undershooting of articulatory targets, and reduced duration of consonant closures. In contrast, Participant Two (P2) demonstrated tongue-palate contact patterns indicative of impaired lingual control in the presence of both normal and increased articulatory segment durations. Potential reasons for the differing articulatory deficits were hypothesized. The current study demonstrated that assessment with EPG identified potential causes of consonant imprecision in two individuals with hypokinetic dysarthria. Directions for speech pathology intervention, salient from the results of the study, were also noted.

Use of electropalatography in the treatment of disordered articulation following traumatic brain injury: A case study

Electropalatography (EPG) was used as a biofeedback tool in a case study of a 30-year-old male with disordered articulation following traumatic brain injury (TBI). Based on qualitative measures of the participant's intelligibility, improved articulation of the fricatives /s/ and /integral/ were selected as treatment targets. Therapy was administered three times a week for 5 weeks. Results showed that word and sentence intelligibility increased approximately 10%, and error patterns for lingual articulation indicated that fricative -> stop and other fricative errors decreased considerably. EPG measures for /s/ exhibited a significantly more anterior main focus of articulatory contact post therapy. Consonant durations were significantly longer during weeks 3 and 4, and this finding was associated with the emergence of an articulatory contact pattern with a groove rather than complete closure. This articulatory pattern appeared inconsistently and was found to vary across articulations of /s/ but also within a single consonant production. For /integral/, the amount of contact was significantly reduced post therapy and an increase in duration was noted during week 4, similar to that occurring in the production of /s/. Spatial and timing measures were more variable than in normal speakers of English and indicated a general increase in variability across weeks for both /s/ and /integral/. It was concluded that, although the correct fricative patterns appeared only intermittently during production of the consonants, there seemed to be sufficient information for the listener to be able to classify the sound as a fricative. As a part of an intervention program, visual EPG biofeedback therapy would appear to have a definite role in assisting dysarthric speakers exhibiting difficulties with lingual articulation in understanding their errors, learning how to exploit kinesthetic, and acoustic sources of feedback, and how to make appropriate adjustments in tongue articulation to increase the level of speech intelligibility.

Speech production in Parkinson's disease: I. An electropalatographic investigation of tongue-palate contact patterns

Previous studies have indicated that consonant imprecision in Parkinson's disease (PD) may result from a reduction in the amplitude of lingual movements or articulatory undershoot. While this has been postulated, direct measurement of the tongue's contact with the hard palate during speech production has not been undertaken. Therefore, the present study aimed to use electropalatography (EPG) to determine the exact nature of tongue-palate contact in a group of individuals with PD and consonant imprecision (n=9). Furthermore, the current investigation also aimed to compare the results of the participants with PD to a group of aged (n=7) and young (n=8) control speakers to determine the relative contribution of ageing of the lingual musculature to any articulatory deficits noted. Participants were required to read aloud the phrase 'I saw a ___ today' with the artificial palate in-situ. Target words included the consonants /l/, /s/ and /t/ in initial position in both the /i/ and /a/ vowel environments. Phonetic transcription of phoneme productions and description of error types was completed. Furthermore, representative frames of contact were employed to describe the features of tongue-palate contact and to calculate spatial palatal indices. Results of the perceptual investigation revealed that perceived undershooting of articulatory targets distinguished the participant group with PD from the control groups. However, objective EPG assessment indicated that undershooting of the target consonant was not the cause of the perceived articulatory errors. It is, therefore, possible that reduced pressure of tongue contact with the hard palate, sub-lingual deficits or impaired articulatory timing resulted in the perceived undershooting of the target consonants.

Speech production in Parkinson's disease: II. Acoustic and electropalatographic investigation of sentence, word and segment durations

Previous investigations employing electropalatography (EPG) have identified articulatory timing deficits in individuals with acquired dysarthria. However, this technology is yet to be applied to the articulatory timing disturbance present in Parkinson's disease (PD). As a result, the current investigation aimed to use EPG to comprehensively examine the temporal aspects of articulation in a group of nine individuals with PD at sentence, word and segment level. This investigation followed on from a prior study (McAuliffe, Ward and Murdoch) and similarly, aimed to compare the results of the participants with PD to a group of aged (n=7) and young controls (n=8) to determine if ageing contributed to any articulatory timing deficits observed. Participants were required to read aloud the phrase I saw a ___ today'' with the EPG palate in-situ. Target words included the consonants /1/, /s/ and /t/ in initial position in both the /i/ and /a/ vowel environments. Perceptual investigation of speech rate was conducted in addition to objective measurement of sentence, word and segment duration. Segment durations included the total segment length and duration of the approach, closure/constriction and release phases of EPG consonant production. Results of the present study revealed impaired speech rate, perceptually, in the group with PD. However, this was not confirmed objectively. Electropalatographic investigation of segment durations indicated that, in general, the group with PD demonstrated segment durations consistent with the control groups. Only one significant difference was noted, with the group with PD exhibiting significantly increased duration of the release phase for /1a/ when compared to both the control groups. It is, therefore, possible that EPG failed to detect lingual movement impairment as it does not measure the complete tongue movement towards and away from the hard palate. Furthermore, the contribution of individual variation to the present findings should not be overlooked.

Treating Jarrod: A core vocabulary approach

This paper is the final paper in a special issue of Advances in Speech-Language Pathology. The paper presents an intervention case study of a 7 year old child with severe phonological difficulties described in Holm and Crosbie (2006). Dodd et al (2006) hypothesised that Jarrod had an underlying deficit of generating phonological plans for word production and suggested that Jarrod would benefit from core vocabulary therapy. This paper reports on one block of core vocabulary therapy undertaken with Jarrod. Pre- and post-intervention measures showed that Jarrod made significant progress. His speech became more consistent and his accuracy (percent consonants correct) increased.