Acoustic Speech Signal Research Articles

Prelingually Deaf Children With Cochlear Implants Show Better Perception of Voice Cues and Speech in Competing Speech Than Postlingually Deaf Adults With Cochlear Implants.

Postlingually deaf adults with cochlear implants (CIs) have difficulties with perceiving differences in speakers' voice characteristics and benefit little from voice differences for the perception of speech in competing speech. However, not much is known yet about the perception and use of voice characteristics in prelingually deaf implanted children with CIs. Unlike CI adults, most CI children became deaf during the acquisition of language. Extensive neuroplastic changes during childhood could make CI children better at using the available acoustic cues than CI adults, or the lack of exposure to a normal acoustic speech signal could make it more difficult for them to learn which acoustic cues they should attend to. This study aimed to examine to what degree CI children can perceive voice cues and benefit from voice differences for perceiving speech in competing speech, comparing their abilities to those of normal-hearing (NH) children and CI adults. CI children's voice cue discrimination (experiment 1), voice gender categorization (experiment 2), and benefit from target-masker voice differences for perceiving speech in competing speech (experiment 3) were examined in three experiments. The main focus was on the perception of mean fundamental frequency (F0) and vocal-tract length (VTL), the primary acoustic cues related to speakers' anatomy and perceived voice characteristics, such as voice gender. CI children's F0 and VTL discrimination thresholds indicated lower sensitivity to differences compared with their NH-age-equivalent peers, but their mean discrimination thresholds of 5.92 semitones (st) for F0 and 4.10 st for VTL indicated higher sensitivity than postlingually deaf CI adults with mean thresholds of 9.19 st for F0 and 7.19 st for VTL. Furthermore, CI children's perceptual weighting of F0 and VTL cues for voice gender categorization closely resembled that of their NH-age-equivalent peers, in contrast with CI adults. Finally, CI children had more difficulties in perceiving speech in competing speech than their NH-age-equivalent peers, but they performed better than CI adults. Unlike CI adults, CI children showed a benefit from target-masker voice differences in F0 and VTL, similar to NH children. Although CI children's F0 and VTL voice discrimination scores were overall lower than those of NH children, their weighting of F0 and VTL cues for voice gender categorization and their benefit from target-masker differences in F0 and VTL resembled that of NH children. Together, these results suggest that prelingually deaf implanted CI children can effectively utilize spectrotemporally degraded F0 and VTL cues for voice and speech perception, generally outperforming postlingually deaf CI adults in comparable tasks. These findings underscore the presence of F0 and VTL cues in the CI signal to a certain degree and suggest other factors contributing to the perception challenges faced by CI adults.

Linguistic modulation of the neural encoding of phonemes.

Speech comprehension entails the neural mapping of the acoustic speech signal onto learned linguistic units. This acousto-linguistic transformation is bi-directional, whereby higher-level linguistic processes (e.g. semantics) modulate the acoustic analysis of individual linguistic units. Here, we investigated the cortical topography and linguistic modulation of the most fundamental linguistic unit, the phoneme. We presented natural speech and "phoneme quilts" (pseudo-randomly shuffled phonemes) in either a familiar (English) or unfamiliar (Korean) language to native English speakers while recording functional magnetic resonance imaging. This allowed us to dissociate the contribution of acoustic vs. linguistic processes toward phoneme analysis. We show that (i) the acoustic analysis of phonemes is modulated by linguistic analysis and (ii) that for this modulation, both of acoustic and phonetic information need to be incorporated. These results suggest that the linguistic modulation of cortical sensitivity to phoneme classes minimizes prediction error during natural speech perception, thereby aiding speech comprehension in challenging listening situations.

Perceptual uncertainty explains activation differences between audiovisual congruent speech and McGurk stimuli.

Face-to-face communication relies on the integration of acoustic speech signals with the corresponding facial articulations. In the McGurk illusion, an auditory /ba/ phoneme presented simultaneously with a facial articulation of a /ga/ (i.e., viseme), is typically fused into an illusory 'da' percept. Despite its widespread use as an index of audiovisual speech integration, critics argue that it arises from perceptual processes that differ categorically from natural speech recognition. Conversely, Bayesian theoretical frameworks suggest that both the illusory McGurk and the veridical audiovisual congruent speech percepts result from probabilistic inference based on noisy sensory signals. According to these models, the inter-sensory conflict in McGurk stimuli may only increase observers' perceptual uncertainty. This functional magnetic resonance imaging (fMRI) study presented participants (20 male and 24 female) with audiovisual congruent, McGurk (i.e., auditory /ba/ + visual /ga/), and incongruent (i.e., auditory /ga/ + visual /ba/) stimuli along with their unisensory counterparts in a syllable categorization task. Behaviorally, observers' response entropy was greater for McGurk compared to congruent audiovisual stimuli. At the neural level, McGurk stimuli increased activations in a widespread neural system, extending from the inferior frontal sulci (IFS) to the pre-supplementary motor area (pre-SMA) and insulae, typically involved in cognitive control processes. Crucially, in line with Bayesian theories these activation increases were fully accounted for by observers' perceptual uncertainty as measured by their response entropy. Our findings suggest that McGurk and congruent speech processing rely on shared neural mechanisms, thereby supporting the McGurk illusion as a valid measure of natural audiovisual speech perception.

Neural tracking measures of speech intelligibility: Manipulating intelligibility while keeping acoustics unchanged.

Neural speech tracking has advanced our understanding of how our brains rapidly map an acoustic speech signal onto linguistic representations and ultimately meaning. It remains unclear, however, how speech intelligibility is related to the corresponding neural responses. Many studies addressing this question vary the level of intelligibility by manipulating the acoustic waveform, but this makes it difficult to cleanly disentangle the effects of intelligibility from underlying acoustical confounds. Here, using magnetoencephalography recordings, we study neural measures of speech intelligibility by manipulating intelligibility while keeping the acoustics strictly unchanged. Acoustically identical degraded speech stimuli (three-band noise-vocoded, ~20 s duration) are presented twice, but the second presentation is preceded by the original (nondegraded) version of the speech. This intermediate priming, which generates a "pop-out" percept, substantially improves the intelligibility of the second degraded speech passage. We investigate how intelligibility and acoustical structure affect acoustic and linguistic neural representations using multivariate temporal response functions (mTRFs). As expected, behavioral results confirm that perceived speech clarity is improved by priming. mTRFs analysis reveals that auditory (speech envelope and envelope onset) neural representations are not affected by priming but only by the acoustics of the stimuli (bottom-up driven). Critically, our findings suggest that segmentation of sounds into words emerges with better speech intelligibility, and most strongly at the later (~400 ms latency) word processing stage, in prefrontal cortex, in line with engagement of top-down mechanisms associated with priming. Taken together, our results show that word representations may provide some objective measures of speech comprehension.

Investigation of self-supervised pre-trained models for classification of voice quality from speech and neck surface accelerometer signals

Prior studies in the automatic classification of voice quality have mainly studied the use of the acoustic speech signal as input. Recently, a few studies have been carried out by jointly using both speech and neck surface accelerometer (NSA) signals as inputs, and by extracting mel-frequency cepstral coefficients (MFCCs) and glottal source features. This study examines simultaneously-recorded speech and NSA signals in the classification of voice quality (breathy, modal, and pressed) using features derived from three self-supervised pre-trained models (wav2vec2-BASE, wav2vec2-LARGE, and HuBERT) and using a support vector machine (SVM) as well as convolutional neural networks (CNNs) as classifiers. Furthermore, the effectiveness of the pre-trained models is compared in feature extraction between glottal source waveforms and raw signal waveforms for both speech and NSA inputs. Using two signal processing methods (quasi-closed phase (QCP) glottal inverse filtering and zero frequency filtering (ZFF)), glottal source waveforms are estimated from both speech and NSA signals. The study has three main goals: (1) to study whether features derived from pre-trained models improve classification accuracy compared to conventional features (spectrogram, mel-spectrogram, MFCCs, i-vector, and x-vector), (2) to investigate which of the two modalities (speech vs. NSA) is more effective as input in the classification task with pre-trained model-based features, and (3) to evaluate whether the deep learning-based CNN classifier can enhance the classification accuracy in comparison to the SVM classifier. The results revealed that the use of the NSA input showed better classification performance compared to the speech signal. Between the features, the pre-trained model-based features showed better classification accuracies, both for speech and NSA inputs compared to the conventional features. The two classifiers performed equally well for all the pre-trained model-based features for both speech and NSA signals. It was also found that the HuBERT features performed better than the wav2vec2-BASE and wav2vec2-LARGE features for both speech and NSA inputs. In particular, when compared to the conventional features, the HuBERT features showed an absolute accuracy improvement of 3%–6% for speech and NSA signals in the classification of voice quality.

Voice Orientation Recognition: New Paradigm of Speech-Based Human-Computer Interaction

As one of the most preferred forms of Human-Computer Interaction (HCI) nowadays, speech-based HCI enables people to communicate verbally with machines, leveraging technologies such as speech recognition and speech synthesis. Current paradigm of speech-based HCI focus on the content of speech only, failing to comprehend deeper pointing information in voice interaction. In particular, when encountering scenarios with multiple smart voice devices around, if people intend to interact with a certain device, the lack of extra pointing information (like the role played by the direction of eye gaze) would cause unintended response from the other devices, resulting in poor interaction experience during HCI. Hence, an interesting problem is: Is it possible for the devices to be aware of the orientation of human voice with only the acoustic speech signals? There is little research studying this topic, except for very a few primary works with much room for improvement. The main challenge of this study lies in capturing the concealed orientation information embedded within the speech signal, while simultaneously maintaining the scheme’s practicality and high precision. In this paper, we propose Oriennet, for identifying the orientation of human voice. With a series of features intentionally designed in view of the indoor voice propagation model and mouth radiation pattern, as well as the application of attention mechanism, Oriennet achieve 95% accuracy in terms of judging whether people are facing the device or not. Even for the fine-grained task of classifying people’s specific orientation from 8 different directions, our work achieved an accuracy of 74%, far outperforming the existed works. We have validated the robustness of Oriennet under various conditions (noisy environment; different people, rooms, languages, locations; fewer microphones), demonstrating its promising applicability in real-life scenarios.

Bilateral Buccal Flap Revision Palatoplasty to Correct Velopharyngeal Dysfunction: Perceptual Speech, Acoustic, and Aerodynamic Outcomes.

The purpose of this study was to analyze perceptual, acoustic, and aerodynamic changes in speech and velopharyngeal function after bilateral buccal flap revision palatoplasty (BBFRP) in patients with repaired cleft palate. Ten consecutive patients ages 4 to 18 years with velopharyngeal dysfunction treated with BBFRP by a single surgeon were evaluated. Using a visual analog scale, nine blinded speech-language pathologists independently rated hypernasality, hyponasality, audible nasal emission, and speech acceptability. Measurements of the acoustic speech signal were used to quantify changes in hypernasality and nasal emission. The pressure flow technique was used to determine changes in velopharyngeal gap size. Complete records were available for eight patients. After surgery, hypernasality decreased ( P < 0.001) and speech acceptability increased ( P < 0.001) significantly. Audible nasal emission was significantly reduced ( P < 0.001). Postoperative acoustic measures showed a reduction of nasal emission and nasalization. Velopharyngeal gap size significantly decreased after BBFRP ( P < 0.001), correlating with lower visual analog scale ratings of hypernasality ( P = 0.015). Hyponasality did not change significantly after surgery ( P = 0.964). No patient developed sleep-disordered breathing. BBFRP resulted in a measurable improvement in hypernasal speech, audible nasal emission, and speech acceptability without significant changes in hyponasality or risk of obstructive sleep apnea. Therapeutic, IV.

Rational speech comprehension: Interaction between predictability, acoustic signal, and noise.

During speech comprehension, multiple sources of information are available to listeners, which are combined to guide the recognition process. Models of speech comprehension posit that when the acoustic speech signal is obscured, listeners rely more on information from other sources. However, these models take into account only word frequency information and local contexts (surrounding syllables), but not sentence-level information. To date, empirical studies investigating predictability effects in noise did not carefully control the tested speech sounds, while the literature investigating the effect of background noise on the recognition of speech sounds does not manipulate sentence predictability. Additionally, studies on the effect of background noise show conflicting results regarding which noise type affects speech comprehension most. We address this in the present experiment. We investigate how listeners combine information from different sources when listening to sentences embedded in background noise. We manipulate top-down predictability, type of noise, and characteristics of the acoustic signal, thus creating conditions which differ in the extent to which a specific speech sound is masked in a way that is grounded in prior work on the confusability of speech sounds in noise. Participants complete an online word recognition experiment. The results show that participants rely more on the provided sentence context when the acoustic signal is harder to process. This is the case even when interactions of the background noise and speech sounds lead to small differences in intelligibility. Listeners probabilistically combine top-down predictions based on context with noisy bottom-up information from the acoustic signal, leading to a trade-off between the different types of information that is dependent on the combination of a specific type of background noise and speech sound.

The timing of visual speech modulates auditory neural processing

In face-to-face communication, visual information from a speaker’s face and time-varying kinematics of articulatory movements have been shown to fine-tune auditory neural processing and improve speech recognition. To further determine whether the timing of visual gestures modulates auditory cortical processing, three sets of syllables only differing in the onset and duration of silent prephonatory movements, before the acoustic speech signal, were contrasted using EEG. Despite similar visual recognition rates, an increase in the amplitude of P2 auditory evoked responses was observed from the longest to the shortest movements. Taken together, these results clarify how audiovisual speech perception partly operates through visually-based predictions and related processing time, with acoustic–phonetic neural processing paralleling the timing of visual prephonatory gestures.

Detection of speech pathology based on parameters of analysis of dysphonia in speech and voice

Objective:To analysis speech pathology based on dysphonia in speech and voice（ADSV）. Methods:The acoustic signals of continuous vowels and continuous speech of one-hundred and thirteen individuals were collected, including 93 vocal cord polyps cases, 20 glottis laryngeal carcinoma cases and 47 volunteers without speech sound disorders. Cepstral peak prominence（CPP）, CPP standard deviation（CPP SD）, L/H spectral ratio（L/H ratio）, L/H ratio standard deviation（L/H ratio SD） and cepstral/spectral index of dysphonia（CSID） were analyzed by ADSV to explore the role of these parameters in the recognition of speech pathology. Results:In the acoustic signal of continuous vowels, CPP and L/H ratio in normal group were higher than those in pathological voice group（P<0.001）, while CPP SD and CSID were lower than those in pathological voice group（P<0.001）, CPP and CSID areas under ROC curve were 0.95 and 0.99, respectively, which were important acoustic parameters for diagnosing pathological voice. In continuous speech acoustic signals, CPP, CPP SD and L/H ratio in the normal group were all higher than those in the speech disorders group（P<0.001）, and the area under the curve of CPP SD was 0.90, which showed high accuracy in diagnosing pathological voice. The ADSV voice analysis parameters CPP, CPP SD, CSID, and L/H ratio also showed significant differences between the vocal cord polyp group and the glottic laryngeal cancer group. The results of the discriminant analysis model show that the use of ADSV voice parameters can distinguish vocal cord polyps and laryngeal cancers. Conclusion:The ADSV voice analysis parameters can not only distinguish the voice signals of the normal group and the pathological group, but also distinguish different types of pathological voices. It has high sensitivity and specificity in diagnosing pathological voices.

Automatic Speech Recognition (ASR) Systems for Children: A Systematic Literature Review

Automatic speech recognition (ASR) is one of the ways used to transform acoustic speech signals into text. Over the last few decades, an enormous amount of research work has been done in the research area of speech recognition (SR). However, most studies have focused on building ASR systems based on adult speech. The recognition of children’s speech was neglected for some time, which means that the field of children’s SR research is wide open. Children’s SR is a challenging task due to the large variations in children’s articulatory, acoustic, physical, and linguistic characteristics compared to adult speech. Thus, the field became a very attractive area of research and it is important to understand where the main center of attention is, and what are the most widely used methods for extracting acoustic features, various acoustic models, speech datasets, the SR toolkits used during the recognition process, and so on. ASR systems or interfaces are extensively used and integrated into various real-life applications, such as search engines, the healthcare industry, biometric analysis, car systems, the military, aids for people with disabilities, and mobile devices. A systematic literature review (SLR) is presented in this work by extracting the relevant information from 76 research papers published from 2009 to 2020 in the field of ASR for children. The objective of this review is to throw light on the trends of research in children’s speech recognition and analyze the potential of trending techniques to recognize children’s speech.

Motor and visual influences on auditory neural processing during speaking and listening

During speaking or listening, endogenous motor or exogenous visual processes have been shown to fine-tune the auditory neural processing of incoming acoustic speech signal. To compare the impact of these cross-modal effects on auditory evoked responses, two sets of speech production and perception tasks were contrasted using EEG. In a first set, participants produced vowels in a self-paced manner while listening to their auditory feedback. Following the production task, they passively listened to the entire recorded speech sequence. In a second set, the procedure was identical except that participants also watched online their own articulatory movements. While both endogenous motor and exogenous visual processes fine-tuned auditory neural processing, these cross-modal effects were found to act differentially on the amplitude and latency of auditory evoked responses. A reduced amplitude was observed on auditory evoked responses during speaking compared to listening, irrespective of the auditory or audiovisual feedback. Adding orofacial visual movements to the acoustic speech signal also speeded up the latency of auditory evoked responses, irrespective of the perception or production task. Taken together, these results suggest distinct motor and visual influences on auditory neural processing, possibly through different neural gating and predictive mechanisms.

Silent speech command word recognition using stepped frequency continuous wave radar

Recovering speech in the absence of the acoustic speech signal itself, i.e., silent speech, holds great potential for restoring or enhancing oral communication in those who lost it. Radar is a relatively unexplored silent speech sensing modality, even though it has the advantage of being fully non-invasive. We therefore built a custom stepped frequency continuous wave radar hardware to measure the changes in the transmission spectra during speech between three antennas, located on both cheeks and the chin with a measurement update rate of 100 Hz. We then recorded a command word corpus of 40 phonetically balanced, two-syllable German words and the German digits zero to nine for two individual speakers and evaluated both the speaker-dependent multi-session and inter-session recognition accuracies on this 50-word corpus using a bidirectional long-short term memory network. We obtained recognition accuracies of 99.17% and 88.87% for the speaker-dependent multi-session and inter-session accuracy, respectively. These results show that the transmission spectra are very well suited to discriminate individual words from one another, even across different sessions, which is one of the key challenges for fully non-invasive silent speech interfaces.

Tongue Postures and Tongue Centers: A Study of Acoustic-Articulatory Correspondences Across Different Head Angles.

Recent research on body and head positions has shown that postural changes may induce varying degrees of changes on acoustic speech signals and articulatory gestures. While the preservation of formant profiles across different postures is suitably accounted for by the two-tube model and perturbation theory, it remains unclear whether it is resulted from the accommodation of tongue postures. Specifically, whether the tongue accommodates the changes in head angle to maintain the target acoustics is yet to be determined. The present study examines vowel acoustics and their correspondence with the articulatory maneuvers of the tongue, including both tongue postures and movements of the tongue center, across different head angles. The results show that vowel acoustics, including pitch and formants, are largely unaffected by upward or downward tilting of the head. These preserved acoustics may be attributed to the lingual gestures that compensate for the effects of gravity. Our results also reveal that the tongue postures in response to head movements appear to be vowel-dependent, and the tongue center may serve as an underlying drive that covariates with the head angle changes. These results imply a close relationship between vowel acoustics and tongue postures as well as a target-oriented strategy for different head angles.

Beyond Recognition: Visual Contributions to Verbal Working Memory.

It is well recognized that adding the visual to the acoustic speech signal improves recognition when the acoustic signal is degraded, but how that visual signal affects postrecognition processes is not so well understood. This study was designed to further elucidate the relationships among auditory and visual codes in working memory, a postrecognition process. In a main experiment, 80 young adults with normal hearing were tested using an immediate serial recall paradigm. Three types of signals were presented (unprocessed speech, vocoded speech, and environmental sounds) in three conditions (audio-only, audio-video with dynamic visual signals, and audio-picture with static visual signals). Three dependent measures were analyzed: (a) magnitude of the recency effect, (b) overall recall accuracy, and (c) response times, to assess cognitive effort. In a follow-up experiment, 30 young adults with normal hearing were tested largely using the same procedures, but with a slight change in order of stimulus presentation. The main experiment produced three major findings: (a) unprocessed speech evoked a recency effect of consistent magnitude across conditions; vocoded speech evoked a recency effect of similar magnitude to unprocessed speech only with dynamic visual (lipread) signals; environmental sounds never showed a recency effect. (b) Dynamic and static visual signals enhanced overall recall accuracy to a similar extent, and this enhancement was greater for vocoded speech and environmental sounds than for unprocessed speech. (c) All visual signals reduced cognitive load, except for dynamic visual signals with environmental sounds. The follow-up experiment revealed that dynamic visual (lipread) signals exerted their effect on the vocoded stimuli by enhancing phonological quality. Acoustic and visual signals can combine to enhance working memory operations, but the source of these effects differs for phonological and nonphonological signals. Nonetheless, visual information can support better postrecognition processes for patients with hearing loss.

Speech Computations of the Human Superior Temporal Gyrus.

Human speech perception results from neural computations that transform external acoustic speech signals into internal representations of words. The superior temporal gyrus (STG) contains the nonprimary auditory cortex and is a critical locus for phonological processing. Here, we describe how speech sound representation in the STG relies on fundamentally nonlinear and dynamical processes, such as categorization, normalization, contextual restoration, and the extraction of temporal structure. A spatial mosaic of local cortical sites on the STG exhibits complex auditory encoding for distinct acoustic-phonetic and prosodic features. We propose that as a population ensemble, these distributed patterns of neural activity give rise to abstract, higher-order phonemic and syllabic representations that support speech perception. This review presents a multi-scale, recurrent model of phonological processing in the STG, highlighting the critical interface between auditory and language systems.

Glottal features for classification of phonation type from speech and neck surface accelerometer signals

Glottal source characteristics vary between phonation types due to the tension of laryngeal muscles with the respiratory effort. Previous studies in the classification of phonation type have mainly used speech signals recorded by microphone. Recently, two studies were published in the classification of phonation type using neck surface accelerometer (NSA) signals. However, there are no previous studies comparing the use of the acoustic speech signal vs. the NSA signal as input in classifying phonation type. Therefore, the current study investigates simultaneously recorded speech and NSA signals in the classification of three phonation types (breathy, modal, pressed). The general goal is to understand which of the two signals (speech vs. NSA) is more effective in the classification task. We hypothesize that by using the same feature set for both signals, classification accuracy is higher for the NSA signal, which is more closely related to the physical vibration of the vocal folds and less affected by the vocal tract compared to the acoustical speech signal. Glottal source waveforms were computed using two signal processing methods, quasi-closed phase (QCP) glottal inverse filtering and zero frequency filtering (ZFF), and a group of time-domain and frequency-domain scalar features were computed from the obtained waveforms. In addition, the study investigated the use of mel-frequency cepstral coefficients (MFCCs) derived from the glottal source waveforms computed by QCP and ZFF. Classification experiments with support vector machine classifiers revealed that the NSA signal showed better discrimination of the phonation types compared to the speech signal when the same feature set was used. Furthermore, it was observed that the glottal features showed complementary information with the conventional MFCC features resulting in the best classification accuracy both for the NSA signal (86.9%) and the speech signal (80.6%).

Prediction and error in early infant speech learning: A speech acquisition model

In the last two decades, statistical clustering models have emerged as a dominant model of how infants learn the sounds of their language. However, recent empirical and computational evidence suggests that purely statistical clustering methods may not be sufficient to explain speech sound acquisition. To model early development of speech perception, the present study used a two-layer network trained with Rescorla-Wagner learning equations, an implementation of discriminative, error-driven learning. The model contained no a priori linguistic units, such as phonemes or phonetic features. Instead, expectations about the upcoming acoustic speech signal were learned from the surrounding speech signal, with spectral components extracted from an audio recording of child-directed speech as both inputs and outputs of the model. To evaluate model performance, we simulated infant responses in the high-amplitude sucking paradigm using vowel and fricative pairs and continua. The simulations were able to discriminate vowel and consonant pairs and predicted the infant speech perception data. The model also showed the greatest amount of discrimination in the expected spectral frequencies. These results suggest that discriminative error-driven learning may provide a viable approach to modelling early infant speech sound acquisition.

Individual Identification Through Voice Using Mel-Frequency Cepstrum Coefficient (MFCC) and Hidden Markov Models (HMM) Method

Voice is one of the parameters in the identification process of a person. Through the voice, information will be obtained such as gender, age, and even the identity of the speaker. Speaker recognition is a method to narrow down crimes and frauds committed by voice. So that it will minimize the occurrence of faking one's identity. The Method of Mel Frequency Cepstrum Coefficient (MFCC) can be used in the speech recognition system. The process of feature extraction of speech signal using MFCC will produce acoustic speech signal. The classification, Hidden Markov Models (HMM) is used to match unidentified speaker’s voice with the voices in database. In this research, the system is used to verify the speaker, namely 15 text dependent in Indonesian. On testing the speaker with the same as database, the highest accuracy is 99,16%.

Categorization of U.S. regional dialects and race from speech

Listeners extract a range of indexical information about speakers from the acoustic speech signal. Listeners show moderate to high accuracy with the two dimensions investigated here—race and regional dialect—with generally higher accuracy for race than regional dialect. In this project, both race and regional dialect identification from the same talkers were tested when the dimensions were varied orthogonally. In a forced-choice categorization task, listeners were presented with 144 unique audio clips of monolingual English talkers from Midland, New York City, and Southern dialect regions who were either Asian, Black, or white. Stimuli were presented to listeners in two separate, counter-balanced testing blocks (i.e., one for race and one for regional dialect). Results showed higher accuracy for race identification than regional dialect, with white Midland talkers most accurately identified by race and Southern Black talkers most accurately identified by dialect. Both tasks showed significant interactions between the two indexical dimensions. The results suggest that an irrelevant indexical dimension can impact perception of the relevant indexical dimension. Further, the composite of a listener's identity and their experiences with people of various ethnicities, racial and dialect groups can shape their perception and response biases. [Work supported by Indiana University Hutton Honors College.]