Framework for Speech Enhancement and Recognition

theoretical account for run-in Enhancement and course credit A Generalized Framework for Speech Enhancement and Recognition with Special Focus On Patients with Speech Disorderslit ReviewKumara Sharma et.al. bemuse proposed Harmonics-to-Noise Ratio and Critical-Band Energy Spectrum of expression communication as acousticalal Indicators of Laryngeal and Voice Pathology 8. Acoustic abbreviation of deliverance call(a) attentions is a noninvasive proficiency that has been proved to be an effective tool for the objective strengthener of straight-from-the-shoulder and articulatio disease screening. In the present study acoustic analysis of keep up vowels is considered. A simple k-means ne argonst neighbor classifier is designed to test the efficacy of a good-hearteds-to- tone proportionality (HNR) measure and the critical-band energy spectrum of the express character communication signal as tools for the sensing of laryngeal pathologies 12. It groups the given voice sign al sample into pathologic and regulation. The soft quarrel signal is decomposed into agreeable and haphazardness components apply an iterative signal extrapolation algorithm. The HNRs at four varied frequency bands argon estimated and used as features. Voiced speech is in any scale leached with 21 critical-band pass filters that mimic the human auditory neurons. Normalized energies of these filter takes are used as another set of features. The HNR and the critical-band energy spectrum whoremonger be used to correlate laryngeal pathology and voice alte dimensionn, using previously classified voice samples. This regularity could be an additional acoustic index number that supplements the clinical diagnostic features for voice evaluation 42.Cepstral-based estimation is used to cater a baseline estimate of the noise level in the logarithmic spectrum for voiced speech. A theoretical description of Cepstral treat of voiced speech containing design noise, together with su pporting empirical data, is provided in order to illustrate the genius of the noise baseline estimation process. Taking the Fourier transform of the liftered (filtered in the Cepstral domain) cepstrum produces a noise baseline estimate. It is shown that Fourier transforming the dispirited-pass liftered cepstrum is comparable to applying a piteous average (MA) filter to the logarithmic spectrum and hence the baseline receives contributions from the glottal cite excited communicative tract and the noise excited outspoken tract43. Because the estimation process resembles the attain of a MA filter, the resulting noise baseline is determined by the kindly resolution as determined by the blase analysis windowpane length and the glottal source spectral tilt. On selecting an appropriate temporal analysis window length the estimated baseline is shown to lie halfway among the glottal excited song tract and the noise excited vocal tract. This information is employed in a new harmo nics-to-noise (HNR) estimation technique, which is shown to provide accurate HNR estimates when tested on synthetically generated voice signals. HNR is defined as the ratio between the energy of the periodic component to the energy of the noncyclic component in the signal. As such it is sensitive to all forms of wave form aperiodicity 8,12. It however specifically reflects a signal to aspiration noise ratio when other aperiodicities in the signal are comparatively low. Validation of a HNR method requires testing the technique against synthesis data with a priori friendship of the HNR.Time-domain methods that require individual period sensing for HNR estimation can be caperatic because of the difficulty in estimating the period markers for pathological voiced speech. frequentlyness domain methods encounter the problem of estimating noise at harmonic locations .Cepstral techniques occupy been introduced to supply noise estimates at all frequency locations in the spectrum (the Cepstral process removes the harmonics from the spectrum).It is shown that the cepstrum-based noise baseline estimation process is comparable to applying a moving average MA filter to the power spectrum and hence the baseline receives contributions from the glottal source excited vocal tract and the noise excited vocal tract. Two crucial issues need to be considered with respect to HNR estimation for sustained vowel phonation when inferring glottal noise levels HNR is a global indicator of voice periodicity.HNR is indirectly related to the noise level of the glottal source .HNR provides a global estimate of signal periodicity. Hence a low value of HNR can arise from any form of aperiodicity, for example, from aspiration noise, jitter, shimmer, nonstationarity of the vocal tract, or other waveform anomalies 43.Daryush Mehta has discussed about Aspiration Noise during vocalization Synthesis, Analysis, and Pitch-Scale Modification. The current study investigates the synthesis and an alysis of aspiration noise in synthesized and spoken vowels. Based on the linear source-filter model of speech production, motive has implemented a vowel synthesizer in which the aspiration noise source is temporally modulated by the periodic source waveform. Modulations in the noise source waveform and their synchronism with the periodic source are shown to be salient for congenital- up justing vowel synthesis. The accurate estimation of the aspiration noise component that contains energy across the frequency spectrum and temporal characteristics due to modulations in the noise source was a challenging task for the author. phantasmal harmonic/noise component analysis of spoken vowels shows evidence of noise modulations with peaks in the estimated noise source component synchronous with both(prenominal) the open contour of the periodic source and with time instants of glottal closure 39.Due to subjective modulations in the aspiration noise source, author has developed an alter nate mount to the speech signal processing with the aim of accurate pitch-scale passing. The proposed strategy takes a dual processing approach, in which the periodic and noise components of the speech signal are separately analyzed, modified, and re-synthesized. The periodic component is modified using our death penalty of time-domain pitch-synchronous overlap-add, and the noise component is handled by modifying characteristics of its source waveform. Author has modeled an internal coupling between the original periodic and aspiration noise sources the modification algorithm is designed to preserve the synchronism between temporal modulations of the devil sources 44. The reconstructed modified signal is perceived to be natural- appear and generally reduces artifacts. Arpit Mathur et.al. produce discussed about the significance of parametric spectral ratio methods in detection and intuition of whispered speech 45.Other ReferencesKaladhar developed confusion ground substance which is a matrix for a two-class classifier, contains information about actual and predicted varietys do by a classification system. The accuracy obtained by knowledge the probabilistic neural network using Parkinson disease dataset got 100% as positives, predictions that an pillow slip is positive, using WEKA 3 and Matlab v7. The data explored in this research was obtained from the Oxford Parkinsons ailment Detection Dataset. Data mining is the process of extracting patterns from data. Data mining is an important tool to transform this data into information. Authors present results with accuracy obtained by training the probabilistic neural network using the higher up dataset 46. Xiao Li et.al. proposed a technique to reduce the likelihood computation in ASR systems that use unceasing density HMMs. Based on the nature of driving features and the numerical properties of Gaussian mixture distributions, the observation likelihood computation is approximated to achieve a speed up. Although the technique does not show appreciable benefit in an isolated vocalise task, it yields significant improvements in continuous speech recognition. For example, 50% of the computation can be saved on the TIMIT database with only a negligible degradation in system performance 47.Authors analyze the case with only soundless features and their deltas and focus on achieving computational saving by go awayially cipher the observation probability in a Gaussian component. It ignores computing the dynamic-feature part of an observation vector when its static-feature part already falls in the drag of a Gaussian. This technique doesnt require a complicated training cognitive process and brings almost no over result to the decoding process. It is effective on both isolated word and connected word speech tasks, but whole kit especially well on connected word recognition with high-dimensional dynamic features 47. Elisabeth Ahlsn has discussed different types of communication disorders. In case of Global aphasia in that location is nil or almost no linguistic communication. In case of Brocas aphasia thither is slow, effortful speech, telegram style, word causeing problems cognize as anomia, relatively good comprehension. In case of Wernickes aphasia there is fluent verbose speech, word finding difficulties known as anomia, substitutions of linguistic communication and sounds, impaired comprehension. In case of Anomic aphasia there are only word finding problems 49.Kristen Jacobson explains about auditory and language processing disorders as follows. There are three general levels that speech sounds travel through with(predicate) art object we are listening. The first level refers to the reception of sounds that occurs inside our ears. A person who is diagnosed with a hearing impairment has difficulties perceiving sounds at this level. This problem is not referred to as a processing disorder. Central auditory processing disorders (CAPD) refer to di fficulties discriminating, identifying and retaining sounds after the ears have heard the sounds. Individuals who recognise difficulties attaching meaning to sound groups that form words, sentences and stories are often diagnosed with language processing disorders. They may excessively experience similar difficulties processing and organizing language for meaning during reading. Similar sounding words are often confused and some individuals may experience sensitivity to specific sounds. Reduced recognition of stress patterns and word boundaries inwardly sentences is often present, especially during rapid speech or listening without opthalmic cues. At times, only parts of messages are received accurately, so that messages and directions often appear incomplete. Specific language processing deficits are often reflected in delayed responses, the need to rehearse statements, and/or the need for frequent reviews plot of ground learning new information 50.There are various types of s peech disorders in tykeren described as follows.Articulation There is difficulty in the production of individual or sequenced sounds. The speakers exhibit substitutions, omissions, additions, and distortions of syllables or words. The Motor or Neurogenic speech disorders result into speech difficulties and affect the preparedness, coordination, timing, and execution of speech movements. Apraxia of speech is neurogenic motor speech disorder affecting the planning of speech. There is difficulty with the voluntary, purposeful movement of speech .The causes are stroke, tumor, head injury, and developmental disorders. The speakers can produce individual sounds but cannot produce them in longer words or sentences. Voice disorders affect pitch, duration, intensity, resonance, and vocal note parameters. Fluency disorders produce interruptions in the flow of speaking. It is withal known as stuttering. It means frequent repetition and/or prolongation of words or sounds 51.Treatment of chi ldren with Speech Oral Placement Disorders (OPD)s needs various types of speech oral placement therapy (OPT) .Children with speech OPDs may have typical or a typical oral structures. The key to the definition of OPD lies in the childs ability or inability to imitate auditory- visual stimuli and follow verbal oral placement instructions. Children with OPD cannot imitate targeted speech sounds using auditory and visual stimuli .They also cannot follow specific instructions to produce targeted speech sounds 52.doubting Thomas Dubuisson et.al. described an analysis system aiming at discriminating between normal and pathological voices. Based on the normal and pathological samples included the MEEI database, it has been name that using two features (spectral decrease and first spectral tristimuli in the scrape scale). Music Information Retrieval (MIR) aims at extracting information from music in order to build classification system of music. Temporal Domain features are Energy, mean, s tandard deviation. Spectral features are spectral Delta, Spectral Mean Value, Spectral Standard Deviation, Spectral Center of Gravity known as spectral centroid, Spectral Moments. The first four importations of the power spectrum M1, M2, M3, M4 . M3 is used to compute the lopsidedness defining the orientation of the PSD around its first moment. If it is positive, the PSD is to a greater extent oriented to the right and to the left if is negative. The skewness is computed as Skewness = M3/(M2)3/2 . The fourth moment is used to compute the kurtosis defining the acuity of the PSD around its first moment. A Gaussian distribution is having a kurtosis equal to 3, a distribution with a higher kurtosis is more acute than a Gaussian one while a distribution with a lower kurtosis is more flat than a Gaussian distribution. The kurtosis is computed asKurtosis = M4/(M2)2. The Soft Phonation Index is defined for the (0 railway yard Hz) and (08000 Hz) frequency bands 54. Behnaz Ghoraani et.al. pr oposed a novel methodology for automatic pattern classification of pathological voices. The main contribution of this paper is extraction of meaningful and eccentric features using Adaptive time-frequency distribution (TFD) and nonnegative matrix factorization (NMF). The proposed method extracts meaningful and unique features from the joint TFD of the speech, and automatically identifies and measures the abnormality of the signal. The proposed method is utilize on the Massachusetts Eye and Ear Infirmary (MEEI) voice disorders database. As a matter of fact from the TFD of abnormal speech it is evident that there are more transients in the abnormal signals, and the formants in pathological speech are more spread and are less structured 55.Corinne Fredouille et.al. have addressed voice disorder assessment. The goal of this methodology is to bring a better understanding of acoustic phenomena related to dysphonia. The automatic system was legalate on dysphonic corpus (80) female voice s. These observations led to a manual analysis of unvoiced plosives, which highlighted a lengthening of VOT according to the dysphonia severity validated by a preliminary statistical analysis. The feature vectors issued from this analysis, at a 10 milli secondly rate, are finally normalized to fit a 0-mean and 1-variance distribution. The LFSC/MFSC computation is through by using the (GPL) SPRO toolkit. Finally, the feature vectors can be augmented by adding dynamic information representing the way these vectors vary in time. Here, first and second derivatives of static coefficients are considered (also named and coefficients) resulting in 72 coefficients 56.Younggwan Kim et.al. discussed the role of the statistical model-based voice activity detector (SMVAD) to detect speech regions from input signals using the statistical models of noise and noisy speech. The LRT-based decision rule may cause detection errors because of statistical properties of noise and speech signals57.Wiqas Ghai et.al. described automatic speech recognition system as comprised of modules Speech Signal acquisition ,Feature extraction, using MFCC is through . Acoustic Modeling is done for expected phonetics of the hypothesis word/sentence. For generating office between the basic speech units such as phones, tri-phones syllables, a fuddled training is carried. During training, a pattern representative for the features of a class using one or more patterns corresponding to speech sounds of the same class. phrase Lexical Modeling is done with the help of Text Corpus, Pronunciation dictionary and vocabulary Model 59.Lucas Leon Oller presents analysis of voice signals for the Harmonics-to-Noise crossover frequency .The harmonics-to-noise ratio (HNR) has been used to assess the behavior of the vocal belt up closure. The objective is to find a particular harmonics-to-noise crossover frequency (HNF) where the harmonic components of the voice repose below the noise floor, and use it as a n indicator of the vocal fold insufficiency. . As the range used for the calculation of the cepstrum approaches the lowest octaves, the growth of the rahmonics should hurry at some stop, the range is going to contain harmonics that are above the noise floor level, and then the energy of the rahmonics will start to faster. That point would be the harmonics-to-noise crossover frequency 60. Daryl Ning has developed an Isolated Word Recognition System in MATLAB. A robust speech-recognition system combines accuracy of identification with the ability to filter out noise and adapt to other acoustic conditions, such as the speakers speech rate and accent. It requires precise knowledge of signal processing and statistical modeling 61.Phonetic ConceptsDaniel Jurafsky et.al. presented a case study of Star trek where robots converse with humans in natural Dialogue system with language conversational agents. Various components that sacrifice up modern conversational agents, including langua ge input and language output dialogue ,automatic speech recognition, natural language understanding ,response planning , speech synthesis systems and the goal of machine translation which leads to automatic translation of a document from one language to another is explained here 62.Steven Pruett describes speech as the motor act of communicating by articulating verbal expression and Language as the knowledge of a symbol system used for social communication. Mary Planchart has explained four domains of language namely Phonology, Grammar , Morphology ,Syntax , and Pragmatics 63, 64.Eric J. Hunter has presented a case study of a 5 year old rose-cheeked male child. He has analyzed comparison of the childs primordial frequencies in structured elicited vocalizations versus unstructured natural vocalizations. The child also wore a National Center for Voice and Speech voice dosimeter, a device that collects voice data over the course of an entire day, during all activities for 34 hours o ver 4 days. It was observed that the childs long-run F0 distribution is not normal. If this distribution is consistent in long-term, unstructured natural vocalization patterns of children, statistical mean would not be a valid measure. Author has suggested mode and median as two parameters which convey more accurate information about typical F0 usage 65.