COMBINED FEATURE EXTRACTION TECHNIQUES AND NAIVE BAYES CLASSIFIER FOR SPEECH RECOGNITION

Speech processing and consequent recognition are important areas of Digital Signal Processing since speech allows people to communicate more natu-rally and efficiently. In this work, a speech recognition system is developed for re-cognizing digits in Malayalam. For recognizing speech, features are to be ex-tracted from speech and hence feature extraction method plays an important role in speech recognition. Here, front end processing for extracting the features is per-formed using two wavelet based methods namely Discrete Wavelet Transforms (DWT) and Wavelet Packet Decomposition (WPD). Naive Bayes classifier is used for classification purpose. After classification using Naive Bayes classifier, DWT produced a recognition accuracy of 83.5% and WPD produced an accuracy of 80.7%. This paper is intended to devise a new feature extraction method which produces improvements in the recognition accuracy. So, a new method called Dis-crete Wavelet Packet Decomposition (DWPD) is introduced which utilizes the hy-brid features of both DWT and WPD. The performance of this new approach is evaluated and it produced an improved recognition accuracy of 86.2% along with Naive Bayes classifier.

Developing a Child Friendly Text-to-Speech System

This paper discusses the implementation details of a child friendly, good quality, English text-to-speech (TTS) system that is phoneme-based, concatenative, easy to set up and use with little memory. Direct waveform concatenation and linear prediction coding (LPC) are used. Most existing TTS systems are unit-selection based, which use standard speech databases available in neutral adult voices.Here reduced memory is achieved by the concatenation of phonemes and by replacing phonetic wave files with their LPC coefficients. Linguistic analysis was used to reduce the algorithmic complexity instead of signal processing techniques. Sufficient degree of customization and generalization catering to the needs of the child user had been included through the provision for vocabulary and voice selection to suit the requisites of the child. Prosody had also been incorporated. This inexpensive TTS systemwas implemented inMATLAB, with the synthesis presented by means of a graphical user interface (GUI), thus making it child friendly. This can be used not only as an interesting language learning aid for the normal child but it also serves as a speech aid to the vocally disabled child. The quality of the synthesized speech was evaluated using the mean opinion score (MOS).

A Study on Futures Trading in Commodities With Special Reference to Rubber

Futures trading in Commodities has three specific economic functions viz. price discovery, hedging and reduction in volatility. Natural rubber possesses all the specifications required for futures trading. Commodity futures trading in India attained momentum after the starting of national level commodity exchanges in 2003. The success of futures trading depends upon effective price risk management, price discovery and reduced volatility which in turn depends upon the volume of trading. In the case of rubber futures market, the volume of trading depends upon the extent of participation by market players like growers, dealers, manufacturers, rubber marketing co-operative societies and Rubber Producer’s Societies (RPS). The extent of participation by market players has a direct bearing on their awareness level and their perception about futures trading. In the light of the above facts and the review of literature available on rubber futures market, it is felt that a study on rubber futures market is necessary to fill the research gap, with specific focus on (1) the awareness and perception of rubber futures market participants viz. (i) rubber growers, (ii) dealers, (iii) rubber product manufacturers, (iv) rubber marketing co-operative societies and Rubber Producer’s Societies (RPS) about futures trading and (2) whether the rubber futures market is fulfilling the economic functions of futures market viz. hedging, reduction in volatility and price discovery or not. The study is confined to growers, dealers, rubber goods manufacturers, rubber marketing co-operative societies and RPS in Kerala. In order to achieve the stated objectives, the study utilized secondary data for the period from 2003 to 2013 from different published sources like bulletins, newsletters, circulars from NMCE, Reserve Bank of India (RBI), Warehousing Corporation and traders. The primary data required for this study were collected from rubber growers, rubber dealers, RPS & Rubber Marketing Co-operative Societies and rubber goods manufacturers in Kerala. Data pertaining to the awareness and perception of futures trading, participation in the futures trading, use of spot and futures prices and source of price information by dealers, farmers, manufacturers and cooperative societies also were collected. Statistical tools used for analysis include percentage, standard deviation, Chi-square test, Mann – Whitney U test, Kruskal Wallis test, Augmented Dickey – Fuller test statistic, t- statistic, Granger causality test, F- statistic, Johansen co – integration test, Trace statistic and Max –Eigen statistic. The study found that 71.5 per cent of the total hedges are effective and 28.5 per cent are ineffective for the period under study. It implies that futures market in rubber reduced the impact of price risks by approximately 71.5 per cent. Further, it is observed that, on 54.4 per cent occasions, the futures market exercised a stabilizing effect on the spot market, and on 45.6 per cent occasions futures trading exercised a destabilizing effect on the spot market. It implies that elasticity of expectation of futures market in rubber has a predominant stabilizing effect on spot prices. The market, as a whole, exhibits a bias in favour of long hedges. Spot price volatility of rubber during futures suspension period is more than that of the pre suspension period and post suspension period. There is a bi-directional association-ship or bi-directional causality or pair- wise causality between spot price and futures price of rubber. From the results of the hedging efficiency, spot price volatility, and price discovery, it can be concluded that rubber futures market fulfils all the economic functions expected from a commodity futures market. Thus in India, the future of rubber futures is Bright…!!!

A diabetes mellitus causa deterioro cognitivo em idosos? Um estudo de revisão

A Diabetes Mellitus (DM) refere-se a uma síndrome metabólica, com prejuízos físicos, sociais e psicológicos naqueles que a possuem. A doença tem um forte componente genético, e é subdividida em Diabetes Mellitus Tipo 1 (DM1), e Diabetes Mellitus Tipo 2 (DM2). A DM2 está associada a déficits funcionais e também cognitivos. Pesquisas revelam que as funções executivas dos idosos com DM encontram-se mais prejudicadas do que naqueles sem DM. Com o objetivo de aprofundar esta temática, esse estudo realiza uma revisão sistemática da literatura, através de publicações indexadas nos últimos oito anos, que abordam a relação entre DM2 e aspectos do deterioro cognitivo em idosos (flexibilidade cognitiva, flexibilidade mental e do pensamento e funções executivas). Nos estudos revisados, a maioria mostrou um grau de comprometimento relacionado com as funções cognitivas: flexibilidade do pensamento, atenção, memória de trabalho, sugerindo, inclusive, que a DM2 acelera o processo de deterioro, aumentando a possibilidade de desenvolvimento de demência. A prevalência de depressão é mais elevada em idosos com DM2, com comprometimento em muitas funções, além de outras complicações físicas identificadas. Os achados apontam para funcionamento cognitivo prejudicado em idosos com DM2, o que enfatiza a necessidade de desenvolvimento de programas de prevenção e intervenção.

Prevalencia de los trastornos de voz ocupacionales en los docentes de dos instituciones educativas de nivel superior en Colombia en el año 2013-2014

Resumen: En la población docente, el uso de la voz como principal herramienta del trabajo por su esfuerzo constante y prolongado constituye el principal factor de riesgo para padecer trastornos de la voz, concomitante con otros factores que inciden en la presentación de estas patologías. Objetivo: Determinar la prevalencia de los trastornos en la voz y sus factores asociados de la población docente de dos instituciones de educación superior (IES) en Colombia y los factores de riesgo asociados. Materiales y métodos: Se realizó un estudio descriptivo de tipo transversal, en el que participaron 149 docentes de dos IES (instituciones de educación superior) de dos ciudades de Colombia. Se aplicó un cuestionario sustentado en el Modelo de Técnicas Foniátricas para Prevenir la Disfonía en Profesores de Secundaria en Portugal y validado por la escuela de Medicina y Ciencias de la Salud, programa de fonoaudiología de la Universidad del Rosario con el trabajo denominado “La Voz como herramienta de trabajo” para Colombia; el cual indagaba los factores riesgo asociados al uso de la voz en el quehacer docente. Se realizó un análisis descriptivo de las variables cualitativas con frecuencias absolutas y porcentajes, medidas de tendencia central, promedio y mediana, de dispersión y desviación estándar. La información recolectada fue analizada con el software SPSS.PASW 18. La correlación entre las variables dependientes y las independientes, se realizaron con el coeficiente no-paramétrico de Spearman y la asociación con la prueba Ji-cuadrado de Pearson o el test exacto de Fisher (valores esperados < 5). Se evaluaron las distribuciones de las variables ordinales de las variables de problemas de la voz con el género, con la prueba no-paramétrica exacta de Mann Whitney. Las pruebas estadísticas se evaluaron a un nivel de significancia del 5% (p<0.05). Resultados: El promedio de edad fue de 36±11.7 años, con un rango que varió entre 25 y 68 años. El género más frecuente fue el masculino con 57,0% de participación. Los datos presentados se dan como resultado de las respuestas a tres preguntas indicativas de los trastornos de voz, las cuales no fueron respondidas por la totalidad de los docentes, por esta razón los datos se presentan así t= número total de personas que respondieron la pregunta y n= número de personas que respondieron afirmativamente. La prevalencia de alteración de voz fue del 63,4% (t=82; n=52), de disminución en el volumen de su voz de 57,3% (t=122; n=70) y cansancio laríngeo en el 76,8% (t=121; n=93). El cansancio laríngeo fue significativamente mayor en los hombres que en las mujeres (p=0.026, Test Mann Whitney exacto a una cola), con algunos hábitos de voz como el realizar esfuerzos en el cuello al hablar (rs = 0,461; p< 0,001) y con la dificultad para inhalar o exhalar aire al hablar (rs = 0,368; p<0.001), con el ruido de fondo (rs =0,361, p<0,001), al permanecer en ambientes contaminados con polvo y humedad (rs=0,311; p=0.001), con sufrir de estrés representado por dolor cervical en espalda y hombros (rs=0,349, p<0.001), y con los problemas familiares (rs= 0,397; p<0.001). Al analizar con la disminución del volumen de voz durante la práctica docente, las correlaciones más significativas mostradas fue estadísticamente mayor entre los hombres que en las mujeres (p=0.006, Test Mann Whitney exacto a una cola), con el esfuerzo cervical para hablar (rs =:0,596, p=<,001), y con el presentar dificultad para inhalar/exhalar aire al usar la voz (rs =0,508, p<,001), con la exposición a ruido de fondo (rs=0,204, p=0,030), ambientes contaminados con polvo y humedad (rs:=305, p<,001), con el estrés (rs =0,316, p=0,001), con los problemas familiares (rs =0,560, p<,001) y en las mujeres con la alteración de la voz en presencia de la menstruación (rs =0,751, p=0,000). Por último, se correlacionó con las alteraciones de voz durante la exposición laboral en la que se mostró mayor correlación en las mujeres que en los hombres (p=0.001, Test Mann Whitney exacto a una cola). Se mostró correlación directa con el no realizar pausas al hablar (rs=0,337, p<0,001), con toser para aclarar la voz (rs =0,349, p=0,001), y con el uso de la voz grave (rs =0,975; p=,004) ,con la presencia de ruido de fondo mientras se dicta clase (rs =0,370, p=0,001) con el permanecer en ambientes contaminados con polvo y humedad (rs =0,253,p=,023), con los síntomas cervicales asociados al estrés (rs=0,323, p=0,003), e inversa con los cambios en tono y volumen de la voz en presencia de menstruación en las mujeres (rs=-0,414,p=,017). Conclusiones: Con los hallazgos encontrados en este estudio es importante reconocer que en las características demográficas, se encontró que el género masculino refirió mayor cantidad de problemas de voz. Para las características laborales la antigüedad ocupacional es factor principal que incide en los problemas de voz; en el análisis de las características demográfico– laborales en las variables analizadas, el grupo etario que reporto cansancio laríngeo se encontraba entre los 36 a 45 años de edad, y los que referían un tiempo de exposición ocupacional mayor a 6 y menor a 15 años. En cuanto a los factores ambientales se identificó que la asociación más significativa está relacionada con estar sometido a cambios bruscos de temperatura, estar expuesto a ruido de fondo durante su labor y finalmente a permanecer en ambientes contaminados. Por ultimo al analizar las alteraciones de la voz con las características personales de los docentes, se observó que el estrés y los problemas familiares inciden de forma significativa en estas patologías. Con estos resultados se recomienda establecer un sistema de vigilancia epidemiológica en las IES, donde se capacite integralmente a los docentes en cuanto a los factores de riesgo personales y de hábitos de voz, además de supervisar y adecuar los factores ambientales que inciden en la presentación de trastornos de voz para evitar la presentación de estas patologías.

A comparison of physical measurements of pure-tones, third-octave bands of noise and third-octave bands of speech to subjective judgments of audibility threshold, MCL and UCL for three normally hearing listeners

This study investigates the relation between physical measurements of pure-tones, third-octave bands of noise and third-octave bands of speech and subjective judgments of auditory threshold, most-comfortable listening level (MCL) and uncomfortable-listening level (UCL) for three normally hearing listeners.

A theme-based creative movement curriculum for deaf and hard of hearing students: Focus on speech, language and audition skills

The purpose of this study was to develop a theme based creative movement curriculum that would help hearing-impaired students develop language, speech and audition skills.

The neural processing of masked speech: Evidence for different mechanisms in the left and right temporal lobes

It has been previously demonstrated that extensive activation in the dorsolateral temporal lobes associated with masking a speech target with a speech masker, consistent with the hypothesis that competition for central auditory processes is an important factor in informational masking. Here, masking from speech and two additional maskers derived from the original speech were investigated. One of these is spectrally rotated speech, which is unintelligible and has a similar (inverted) spectrotemporal profile to speech. The authors also controlled for the possibility of “glimpsing” of the target signal during modulated masking sounds by using speech-modulated noise as a masker in a baseline condition. Functional imaging results reveal that masking speech with speech leads to bilateral superior temporal gyrus (STG) activation relative to a speech-in-noise baseline, while masking speech with spectrally rotated speech leads solely to right STG activation relative to the baseline. This result is discussed in terms of hemispheric asymmetries for speech perception, and interpreted as showing that masking effects can arise through two parallel neural systems, in the left and right temporal lobes. This has implications for the competition for resources caused by speech and rotated speech maskers, and may illuminate some of the mechanisms involved in informational masking.

The neural processing of masked speech: evidence for different mechanisms in the left and right temporal lobes

It has been previously demonstrated that extensive activation in the dorsolateral temporal lobes associated with masking a speech target with a speech masker, consistent with the hypothesis that competition for central auditory processes is an important factor in informational masking. Here, masking from speech and two additional maskers derived from the original speech were investigated. One of these is spectrally rotated speech, which is unintelligible and has a similar (inverted) spectrotemporal profile to speech. The authors also controlled for the possibility of "glimpsing" of the target signal during modulated masking sounds by using speech-modulated noise as a masker in a baseline condition. Functional imaging results reveal that masking speech with speech leads to bilateral superior temporal gyrus (STG) activation relative to a speech-in-noise baseline, while masking speech with spectrally rotated speech leads solely to right STG activation relative to the baseline. This result is discussed in terms of hemispheric asymmetries for speech perception, and interpreted as showing that masking effects can arise through two parallel neural systems, in the left and right temporal lobes. This has implications for the competition for resources caused by speech and rotated speech maskers, and may illuminate some of the mechanisms involved in informational masking.

From dichotic listening to the irrelevant sound effect: a behavioural and neuroimaging analysis of the processing of unattended speech

The assumption that ignoring irrelevant sound in a serial recall situation is identical to ignoring a non-target channel in dichotic listening is challenged. Dichotic listening is open to moderating effects of working memory capacity (Conway et al., 2001) whereas irrelevant sound effects (ISE) are not (Beaman, 2004). A right ear processing bias is apparent in dichotic listening, whereas the bias is to the left ear in the ISE (Hadlington et al., 2004). Positron emission tomography (PET) imaging data (Scott et al., 2004, submitted) show bilateral activation of the superior temporal gyrus (STG) in the presence of intelligible, but ignored, background speech and right hemisphere activation of the STG in the presence of unintelligible background speech. It is suggested that the right STG may be involved in the ISE and a particularly strong left ear effect might occur because of the contralateral connections in audition. It is further suggested that left STG activity is associated with dichotic listening effects and may be influenced by working memory span capacity. The relationship of this functional and neuroanatomical model to known neural correlates of working memory is considered.

Non-word repetition in adolescents with Specific Language Impairment and Autism plus Language Impairments: a qualitative analysis

Non-word repetition (NWR) was investigated in adolescents with typical development, Specific Language Impairment (SLI) and Autism Plus language Impairment (ALI) (n = 17, 13, 16, and mean age 14;4, 15;4, 14;8 respectively). The study evaluated the hypothesis that poor NWR performance in both groups indicates an overlapping language phenotype (Kjelgaard & Tager-Flusberg, 2001). Performance was investigated both quantitatively, e.g. overall error rates, and qualitatively, e.g. effect of length on repetition, proportion of errors affecting phonological structure, and proportion of consonant substitutions involving manner changes. Findings were consistent with previous research (Whitehouse, Barry, & Bishop, 2008) demonstrating a greater effect of length in the SLI group than the ALI group, which may be due to greater short-term memory limitations. In addition, an automated count of phoneme errors identified poorer performance in the SLI group than the ALI group. These findings indicate differences in the language profiles of individuals with SLI and ALI, but do not rule out a partial overlap. Errors affecting phonological structure were relatively frequent, accounting for around 40% of phonemic errors, but less frequent than straight Consonant-for-Consonant or vowel-for-vowel substitutions. It is proposed that these two different types of errors may reflect separate contributory mechanisms. Around 50% of consonant substitutions in the clinical groups involved manner changes, suggesting poor auditory-perceptual encoding. From a clinical perspective algorithms which automatically count phoneme errors may enhance sensitivity of NWR as a diagnostic marker of language impairment. Learning outcomes: Readers will be able to (1) describe and evaluate the hypothesis that there is a phenotypic overlap between SLI and Autism Spectrum Disorders (2) describe differences in the NWR performance of adolescents with SLI and ALI, and discuss whether these differences support or refute the phenotypic overlap hypothesis, and (3) understand how computational algorithms such as the Levenshtein Distance may be used to analyse NWR data.

Cancellation of siren noise from two way voice communications inside emergency vehicles

Sirens used by police, fire and paramedic vehicles have been designed so that they can be heard over large distances, but unfortunately the siren noise enters the vehicle and corrupts intelligibility of voice communications from the emergency vehicle to the control room. Often the siren needs to be turned off to enable the control room to hear what is being said. This paper discusses a siren noise filter system that is capable of removing the siren noise picked up by the two-way radio microphone inside the vehicle. The removal of the siren noise improves the response time for emergency vehicles and thus save lives. To date, the system has been trialed within a fire tender in a non-emergency situation, with good results.

Parenting by anxious mothers: effects of disorder subtype, context and child characteristics

Background:  There has been increasing research interest in parenting by anxious adults; however, little is known about anxiety-subtype effects, or effects of the context in which parenting is assessed. Methods:  Two groups of anxious mothers, social phobia (N = 50), generalised anxiety disorder (N = 38), and nonanxious controls (N = 62) were assessed with their 4.9-year-old children in three tasks: two presented threat specifically relevant to each maternal disorder, namely, a social threat task where the child had to give a speech, and a nonsocial threat task where the child had to explore potentially scary objects; the third was a nonthreat task (playing with play dough). Seven parenting dimensions were scored. Effects on parenting of maternal anxiety subgroup and task, and their interactions, were examined, as were effects of earlier child behavioural inhibition and currently manifest anxiety. Results:  There were no parenting differences between maternal groups in the nonthreat play-dough task; parenting difficulties in the two anxious groups were principally evident in the disorder-specific challenge. Parenting differences between nonanxious and anxious mothers occurred independently of child characteristics. There was little evidence for particular forms of parenting difficulty being unique to maternal disorder. Conclusions:  Anxious mothers’ parenting difficulties emerge when occurring under challenge, especially when this is disorder-specific. These effects should be considered in research and clinical practice.