Segmentation Of Continuous Speech Research Articles

Unsupervised phoneme segmentation of continuous Arabic speech

Unsupervised phoneme segmentation of continuous Arabic speech

The Early Subcortical Response at the Fundamental Frequency of Speech Is Temporally Separated from Later Cortical Contributions.

Most parts of speech are voiced, exhibiting a degree of periodicity with a fundamental frequency and many higher harmonics. Some neural populations respond to this temporal fine structure, in particular at the fundamental frequency. This frequency-following response to speech consists of both subcortical and cortical contributions and can be measured through EEG as well as through magnetoencephalography (MEG), although both differ in the aspects of neural activity that they capture: EEG is sensitive to both radial and tangential sources as well as to deep sources, whereas MEG is more restrained to the measurement of tangential and superficial neural activity. EEG responses to continuous speech have shown an early subcortical contribution, at a latency of around 9 msec, in agreement with MEG measurements in response to short speech tokens, whereas MEG responses to continuous speech have not yet revealed such an early component. Here, we analyze MEG responses to long segments of continuous speech. We find an early subcortical response at latencies of 4-11 msec, followed by later right-lateralized cortical activities at delays of 20-58 msec as well as potential subcortical activities. Our results show that the early subcortical component of the FFR to continuous speech can be measured from MEG in populations of participants and that its latency agrees with that measured with EEG. They furthermore show that the early subcortical component is temporally well separated from later cortical contributions, enabling an independent assessment of both components toward further aspects of speech processing.

Continuous speech segmentation by L1 and L2 speakers of English: the role of syntactic and prosodic cues

We investigated how native and fluent users of English segment continuous speech and to what extent they use sound-related and structure-related cues. As suggested by the notion of multi-competence, L1 users are not seen as ideal models with perfect command of English, and L2 users not as lacking in competence. We wanted to see how language experience affects speech segmentation. We had participants listen to extracts of spontaneous spoken English and asked them to mark boundaries between speech segments. We found that in chunking authentic speech, both groups made the most use of prosody, with L1 users relying slightly more on it. However, the groups did not differ in segmentation strategies and performed alike in efficiency and agreement. Results show that in line with multi-competence, native speakers do not have an advantage over fluent speakers in higher-level speech processes, and that the outcome of natural speech comprehension does not significantly depend on the different language experiences at high levels of fluency. We suggest that research on speech segmentation should take natural continuous speech on board and investigate fluent users independently of their L1s to gain a more holistic view of processes and consequences of speech segmentation.

An Efficient Syllable-Based Speech Segmentation Model Using Fuzzy and Threshold-Based Boundary Detection

To develop a high-quality TTS system, an appropriate segmentation of continuous speech into the syllabic units plays a vital role. The significant objective of this research work involves the implementation of an automatic syllable-based speech segmentation technique for continuous speech of the Hindi language. Here, the parameters involved in the segmentation process are optimized to segment the speech syllables. In addition to this, the proposed iterative splitting process containing the optimum parameters minimizes the deletion errors. Thus, the optimized iterative incorporation can discard more insertions without merging the frequent non-iterative incorporation. The mixture of optimized iterative and iterative incorporation provides the best accuracy with the least insertion and deletion errors. The segmentation output based on different text signals for the proposed approach and other techniques namely GA, PSO and SOM is accurately segmented. The average accuracy obtained for the proposed approach is high with 97.5% than GA, PSO and SOM. The performance of the proposed algorithm is also analyzed and gives better-segmented accuracy when compared with other state-of-the-art methods. Here, the syllable-based segmented database is suitable for the speech technology system for Hindi in the travel domain.

Automatic speech recognition by using local adaptive thresholding in continuous speech segmentation

Speech recognition of continuous speech will be greatly influenced by the word segmentation process of the speech input. Proper segmentation will result in better speech recognition. This study proposed automatic speech recognition by applying local adaptive thresholding in the segmentation process. The segmentation method used is an enhanced blocking block area method whose input is a spectrogram image of the speech signal. While the locally adaptive thresholding method used is the Niblack method which is the best method compared to other methods, namely Sauvola, Bradley, Guanglei Xiong, and Bernsen when applied to the enhanced blocking block area method. For the speech recognition process, using Mel-frequency cepstral coefficients (MFCC) as a feature extraction method and Hidden Markov Model (HMM) as speech recognition. The experimental results show that by using 400 sentences consisting of 80 testing data and 320 as training data and using K-fold cross-validation, the highest accuracy is 60,8%. This result has no significant difference with the use of global thresholding in the segmentation process.

A longitudinal study of automated analysis of acoustic speech markers in FTD and PPA

AbstractBackgroundSpeech is a complex activity requiring proper function and connectivity of multiple brain networks and as such is sensitive to focal neurodegeneration. We have previously reported on acoustic markers of dysprosody in speech samples of speakers with frontotemporal dementia (FTD) phenotypes. In the current study we explore the longitudinal changes in acoustic‐prosodic markers in FTD.MethodWe analyzed 102 speech samples of picture descriptions from 46 participants with FTD (Table 1): 8 with non‐fluent/agrammatic primary progressive aphasia (naPPA), 14 with semantic variant PPA (svPPA), 10 with logopenic aphasia (lvPPA) and 14 with behavioral FTD (bvFTD). We automatically segmented the acoustic signal into segments of continuous speech or silence, measured their durations, and derived other measures. We used linear mixed effects (lme) models to test changes over time for each acoustic measure, controlling for sex, education, and random intercepts. We also examined any interaction between phenotypes and disease duration.ResultbvFTD speakers increased their pause duration by 0.27 seconds per year and their pause rate by 3.9 pauses per minute (ppm) each year. Their speech segment duration shortened by 0.1 seconds per year (p=0.041), decreasing their total speech time by 6.6 seconds (p=0.003) per year. Thus, bvFTD patients reduced the proportion of speech in their samples by 5.16 percent per year (p=0.008). svPPA speakers increased their pause rate similarly, but in contrast, their pause duration decreased by 0.097 seconds per year and they increased their speech segment frequency by 8.32 per minute each year (p=0.054). naPPA and lvPPA speakers increased their pause rate over time and spent less total time (speech + pause) describing the picture (by 5.6 seconds per year; p=0.018). They did not differ from bvFTD and svPPA in these two acoustic measures.ConclusionIn our study all FTD speakers became more dysfluent and produced shorter descriptions with time, however, only bvFTD speakers actually exhibited reduced speech production. In contrast, svPPA speech had more frequent pauses and speech segments over time, rendering it “fragmented” and inefficient. These findings support the role of automated acoustic analysis in characterizing speech longitudinally in neurodegeneration.

Enhanced blocking block area method for segmentation of continuous speech

Segmentation is one of the important stages in the speech recognition in the type of continuous speech. The segmentation phase serves to break the sentences into words that can be recognized by the computer. The quality of segmentation results can affect the results of the recognition that is done. This research examines the dynamic threshold used in the process of continuous speech segmentation and also proposes an Enhanced Blocking Block Area method in the Indonesian language domain. Three algorithms were compared (K-Means, Fuzzy C-Means, and Otsu) to find the best dynamic threshold and add morphological operations and overlapping to the blocking block area method to obtain the best segmentation accuracy. Based on the results of the research, the Fuzzy C-Means algorithm provides the best threshold results compared to the other two algorithms. By using the Fuzzy C-Means algorithm with the addition of morphology and overlapping, this study can improve the accuracy of continuous speech segmentation in the Indonesian Language from 24% to 90%.

A genetic model for acoustic and phonetic decoding of standard arabic vowels in continuous speech

This paper presents an Acoustic and Phonetic Decoding Model (APDM) for automatic recognition Standard Arab (S.A) plain (pure) and emphatic vowels sounds of continuous naturally spoken speech, using Genetic Algorithms (G.As). SA vowels were selected since they are the most difficult phonemes to recognize. We have used GAs because of their advantages in resolving complicated optimization problems and because the results are obtained more rapidly than in the classical methods. In addition, the computational cost is greatly reduced. Our Genetic APDM performs automatically and in parallel, the operation of concatenations of short-term parametric vectors during the speech continuum segmentation stage, and the classification of the acoustic segments of continuous and natural speech into different vowel classes. In order to perform our task, we have used both the Mel Frequency Cepstrum Coding (MFCC) and the Linear Prediction Coding (LPC) methods to extract vocal tract parametric coefficients from the speech signal successfully. Among a set of classifiers we have used the distance one. This paper explains how we have used the Manhattan distance as decision rule the GA evaluation to classify the discriminate parameters vectors. The analysed corpus contains hundreds of sentences composed of the all types of SA vowels in different contexts and recorded by several Algerian male and female speakers, in quite noisy environment. The Corpus phonemes were classified successfully with an overall average rate of 98.02%.

Continuous speech segmentation using local adaptive thresholding technique in the blocking block area method

Continuous speech is a form of natural human speech that is continuous without a clear boundary between words. In continuous speech recognition, a segmentation process is needed to cut the sentence at the boundary of each word. Segmentation becomes an important step because a speech can be recognized from the word segments produced by this process. The segmentation process in this study was carried out using local adaptive thresholding technique in the blocking block area method. This study aims to conduct performance comparisons for five local adaptive thresholding methods (Niblack, Sauvola, Bradley, Guanglei Xiong and Bernsen) in continuous speech segmentation to obtain the best method and optimum parameter values. Based on the results of the study, Niblack method is concluded as the best method for continuous speech segmentation in Indonesian language with the accuracy value of 95%, and the optimum parameter values for such method are window = 75 and k = 0.2.

Automatic Segmentation of Hindi Speech into Syllable-Like Units

To develop the high-quality Text-to-Speech (TTS) system, appropriate segmentation of continuous speech into the syllabic units placed an important role. The research work has been implemented for automatic syllable based speech segmentation technique for continuous speech for the Hindi language. The experiments were conducted by using the energy convex hull approach for clean, continuous speech for Hindi. In this method, the Savitzky-Golay filter was applied on the short term energy (STE) signal to increase the signal to noise ratio (SNR), followed by applying the median filter to preserve the boundaries, hence smoothing the energy curve. Also, the Hamming sliding-window was applied twice on speech signal to get the more accurate depth of convex hull valleys. Further, the algorithm was tested on 50 unique utterances chosen from the travel domain. The accuracy of the proposed algorithm has been calculated and obtains that 76.07% syllables have time-error less than 30 ms with manual segmentation reference. The performance of the proposed algorithm is also analyzed and gives better-segmented accuracy as compared to the existing group delay segmentation technique for fricatives or nasal sounds. The syllable base segmented database is suitable for the speech technology system for Hindi in the travel domain.

Robust neuronal oscillatory entrainment to speech displays individual variation in lateralisation

ABSTRACTNeural oscillations may be instrumental for the tracking and segmentation of continuous speech. Earlier work has suggested that delta, theta and gamma oscillations entrain to the speech rhythm. We used magnetoencephalography and a large sample of 102 participants to investigate oscillatory entrainment to speech, and observed robust entrainment of delta and theta activity, and weak group-level gamma entrainment. We show that the peak frequency and the hemispheric lateralisation of the entrainment are subject to considerable individual variability. The first finding may support the involvement of intrinsic oscillations in entrainment, and the second finding suggests that there is no systematic default right-hemispheric bias for processing acoustic signals on a slow time scale. Although low frequency entrainment to speech is a robust phenomenon, the characteristics of entrainment vary across individuals, and this variation is important for understanding the underlying neural mechanisms of entrainment, as well as its functional significance.

The speech signal segmentation algorithm using pitch synchronous analysis

Abstract Parameterization of the speech signal using the algorithms of analysis synchronized with the pitch frequency is discussed. Speech parameterization is performed by the average number of zero transitions function and the signal energy function. Parameterization results are used to segment the speech signal and to isolate the segments with stable spectral characteristics. Segmentation results can be used to generate a digital voice pattern of a person or be applied in the automatic speech recognition. Stages needed for continuous speech segmentation are described.

Fuzzy-based algorithm for Fongbe continuous speech segmentation

Text-independent speech segmentation is a challenging topic in computer-based speech recognition systems. This paper proposes a novel time-domain algorithm based on fuzzy knowledge for continuous speech segmentation task via a nonlinear speech analysis. Short-term energy, zero-crossing rate and the singularity exponents are the time-domain features that we have calculated in each point of speech signal in order to exploit relevant information for generating the significant segments. This is down for the phoneme or syllable identification and the transition fronts. Fuzzy logic technique helped us to fuzzify the calculated features into three complementary sets namely: low, medium, high and to perform a matching phase using a set of fuzzy rules. The outputs of our proposed algorithm are silence, phonemes, or syllables. Once evaluated, our algorithm produced the best performances with efficient results on Fongbe language (an African tonal language spoken especially in Benin, Togo and Nigeria).

Seeking Temporal Predictability in Speech: Comparing Statistical Approaches on 18 World Languages.

Temporal regularities in speech, such as interdependencies in the timing of speech events, are thought to scaffold early acquisition of the building blocks in speech. By providing on-line clues to the location and duration of upcoming syllables, temporal structure may aid segmentation and clustering of continuous speech into separable units. This hypothesis tacitly assumes that learners exploit predictability in the temporal structure of speech. Existing measures of speech timing tend to focus on first-order regularities among adjacent units, and are overly sensitive to idiosyncrasies in the data they describe. Here, we compare several statistical methods on a sample of 18 languages, testing whether syllable occurrence is predictable over time. Rather than looking for differences between languages, we aim to find across languages (using clearly defined acoustic, rather than orthographic, measures), temporal predictability in the speech signal which could be exploited by a language learner. First, we analyse distributional regularities using two novel techniques: a Bayesian ideal learner analysis, and a simple distributional measure. Second, we model higher-order temporal structure—regularities arising in an ordered series of syllable timings—testing the hypothesis that non-adjacent temporal structures may explain the gap between subjectively-perceived temporal regularities, and the absence of universally-accepted lower-order objective measures. Together, our analyses provide limited evidence for predictability at different time scales, though higher-order predictability is difficult to reliably infer. We conclude that temporal predictability in speech may well arise from a combination of individually weak perceptual cues at multiple structural levels, but is challenging to pinpoint.

AAYUDHA: A Tool for Automatic Segmentation and Labelling of Continuous Tamil Speech

An effective way of communication between human is now becoming an alternative way to communicate between human and machine. This alternative way is now-a-days used in many real time systems for faster, easier and comfortable response and communication. Speech segmentation and labelling are the process that lay as a key to decide the accuracy of several speech related research. A tool AAYUDHA is proposed that enables automatic segmentation and labelling of continuous speech in Tamil. Two different segmentation algorithms, one based on Fast Fourier Transform (FFT) feature set and 2D filtering and other based on Discrete Wavelet Transform (DWT) feature set and its energy variation in different sub-bands are implemented. The segmentation accuracy of those algorithms is analyzed. Further the segmented speech is labelled using a baseline Hidden Markov Model (HMM) based acoustic model. A speech corpus named KAZHANGIYAM is created which includes the recorded Tamil speech of various speakers. The database also includes the information of manually segmented data of those speech data. This speech corpus is used to analyze the accuracy of the algorithms used in the proposed tool. This tool concentrates on the phonetic level segmentation of Tamil speech. The tool shows an acceptable segmentation and labelling accuracy.

Automatic continuous speech recogniser for Dravidian languages using the auto associative neural network

In recent times with the extensive improvement of computers, numerous methods of data interchange between man and computer are revealed. It aims to provide an efficient way for human to communicate with computers exclusively for people with disabilities who face diversity of obstacles while using computers. This paper predominantly focuses on developing an efficient speech recognition system for Dravidian languages such as Tamil, Malayalam, Telugu and Kannada. The proposed CSR system comprises of four steps namely pre-processing, feature extraction, automatic continuous speech segmentation and classification. The most powerful and widely used short term energy and zero crossing rate is used for continuous speech segmentation and Mel frequency cepstral coefficients MFCC, linear predictive cepstral coefficients LPCC and shifted delta cepstrum SDC feature extractions are used for recognition system. Experiments are carried out with real time Dravidian languages speech signal. It is observed from the results that the proposed system gives significant results in AANN classifier with MFCC feature when compared with LPCC and SDC features.

Segmentation of Continuous Tamil Speech into Syllable like Units

The present growth in the field of information and communication technologies has diverted the focus of many researchers towards the speech technologies. Speech technology comprises of many subfields like speech synthesis, speech recognition, speaker recognition, speech compression, speaker verification and Multimodal interaction. The basic units of the speech synthesis and speech recognition system are syllable, phoneme and word. This study mainly focuses on syllable segmentation or syllabification with the aim to further develop a speech synthesis tool in Tamil language for Human Computer Interaction [HCI]. The syllable boundaries are identified using the formant frequency, F1. The proposed syllable segmentation algorithm is applied and tested on a set of recorded continuous speech corpus. Initially, the continuous speech signal is divided into segments by removing the silence regions. The silence removal method used in this work depends on features such as signal energy and spectral centroid. After removing silence portion from the speech signals, the speech segments are further processed using Linear Predictive Coding (LPC) to extract the formant frequencies. Then the peaks in the formant frequencies are used as clue to mark the syllable boundaries in the speech. The proposed algorithm is producing an average accuracy of 89% in identifying syllable boundaries when it is compared with the hand labeled syllable boundaries.

Blocking Black Area Method for Speech Segmentation

Speech segmentation is an important sub problem of automatic speech recognition. This research is concerned with the development of a continuous speech segmentation system using Bangla Language. This paper presents a dynamic thresholding algorithm to segment the continuous Bngla speech sentences into words/sub-words. The research uses Otsu’s method for dynamic thresholding and introduces a new approach, named blocking black area method to identify the voiced regions of the continuous speech in speech segmentation. The developed system has been justified with continuously spoken several Bangla sentences. To test the performance of the system, 100 Bangla sentences have been recorded from 5 (five) male speakers of different ages and 656 words have been presented in the 100 Bangla sentences. So, the speech database contains 500 Bangla sentences with 3280 words. All the algorithms and methods used in this research are implemented in MATLAB and the proposed system has been achieved the average segmentation accuracy of 90.58%.

Speech Rhythms and Multiplexed Oscillatory Sensory Coding in the Human Brain

Author SummaryContinuous speech is organized into a nested hierarchy of quasi-rhythmic components (prosody, syllables, phonemes) with different time scales. Interestingly, neural activity in the human auditory cortex shows rhythmic modulations with frequencies that match these speech rhythms. Here, we use magnetoencephalography and information theory to study brain oscillations in participants as they process continuous speech. We show that auditory brain oscillations at different frequencies align with the rhythmic structure of speech. This alignment is more precise when participants listen to intelligible rather than unintelligible speech. The onset of speech resets brain oscillations and improves their alignment to speech rhythms; it also improves the alignment between the different frequencies of nested brain oscillations in the auditory cortex. Since these brain oscillations reflect rhythmic changes in neural excitability, they are strong candidates for mediating the segmentation of continuous speech at different time scales corresponding to key speech components such as syllables and phonemes.

Transition Probabilities and Different Levels of Prominence in Segmentation

A large body of empirical research demonstrates that people exploit a wide variety of cues for the segmentation of continuous speech in artificial languages, including rhythmic properties, phrase boundary cues, and statistical regularities. However, less is known regarding how the different cues interact. In this study we addressed the question of the relative importance of lexical stress, phrasal prominence, and transitional probabilities (TP) between adjacent syllables for the segmentation of an artificial language. We explored how duration increase, pitch rise, and the combination of duration and pitch on the antepenultimate, the penultimate, and the final syllable of a three‐syllabic word affect segmentation by native speakers of Italian. Our results indicate that, if the most frequent location of stress in the participants’ native language and a lengthened syllable in the artificial language do not coincide, segmentation is disrupted. If there is no conflict between the location of stress in the native language of the participant and the lengthened syllable in the artificial language, segmentation is neither impeded nor facilitated. Pitch marked the edges of the TP‐defined words in a continuous speech stream. When TPs and pitch cues are in conflict, segmentation fails; if pitch rise coincides with the edges of TP words, segmentation succeeds, but is not facilitated. Phrasal prominence comprising both pitch and duration facilitates segmentation when aligned with the word edges. Our findings show that language‐specific peculiarities of how nuclear pitch accents are realized in the native language of the listener might interact with statistical cues in the segmentation of an unfamiliar language.