Matching Items (26)
Filtering by
- All Subjects: Computer Science
- All Subjects: Speech therapy
- Resource Type: Text
Description
Everyday speech communication typically takes place face-to-face. Accordingly, the task of perceiving speech is a multisensory phenomenon involving both auditory and visual information. The current investigation examines how visual information influences recognition of dysarthric speech. It also explores where the influence of visual information is dependent upon age. Forty adults participated in the study that measured intelligibility (percent words correct) of dysarthric speech in auditory versus audiovisual conditions. Participants were then separated into two groups: older adults (age range 47 to 68) and young adults (age range 19 to 36) to examine the influence of age. Findings revealed that all participants, regardless of age, improved their ability to recognize dysarthric speech when visual speech was added to the auditory signal. The magnitude of this benefit, however, was greater for older adults when compared with younger adults. These results inform our understanding of how visual speech information influences understanding of dysarthric speech.
ContributorsFall, Elizabeth (Author) / Liss, Julie (Thesis advisor) / Berisha, Visar (Committee member) / Gray, Shelley (Committee member) / Arizona State University (Publisher)
Created2014
Description
Audio signals, such as speech and ambient sounds convey rich information pertaining to a user’s activity, mood or intent. Enabling machines to understand this contextual information is necessary to bridge the gap in human-machine interaction. This is challenging due to its subjective nature, hence, requiring sophisticated techniques. This dissertation presents a set of computational methods, that generalize well across different conditions, for speech-based applications involving emotion recognition and keyword detection, and ambient sounds-based applications such as lifelogging.
The expression and perception of emotions varies across speakers and cultures, thus, determining features and classification methods that generalize well to different conditions is strongly desired. A latent topic models-based method is proposed to learn supra-segmental features from low-level acoustic descriptors. The derived features outperform state-of-the-art approaches over multiple databases. Cross-corpus studies are conducted to determine the ability of these features to generalize well across different databases. The proposed method is also applied to derive features from facial expressions; a multi-modal fusion overcomes the deficiencies of a speech only approach and further improves the recognition performance.
Besides affecting the acoustic properties of speech, emotions have a strong influence over speech articulation kinematics. A learning approach, which constrains a classifier trained over acoustic descriptors, to also model articulatory data is proposed here. This method requires articulatory information only during the training stage, thus overcoming the challenges inherent to large-scale data collection, while simultaneously exploiting the correlations between articulation kinematics and acoustic descriptors to improve the accuracy of emotion recognition systems.
Identifying context from ambient sounds in a lifelogging scenario requires feature extraction, segmentation and annotation techniques capable of efficiently handling long duration audio recordings; a complete framework for such applications is presented. The performance is evaluated on real world data and accompanied by a prototypical Android-based user interface.
The proposed methods are also assessed in terms of computation and implementation complexity. Software and field programmable gate array based implementations are considered for emotion recognition, while virtual platforms are used to model the complexities of lifelogging. The derived metrics are used to determine the feasibility of these methods for applications requiring real-time capabilities and low power consumption.
The expression and perception of emotions varies across speakers and cultures, thus, determining features and classification methods that generalize well to different conditions is strongly desired. A latent topic models-based method is proposed to learn supra-segmental features from low-level acoustic descriptors. The derived features outperform state-of-the-art approaches over multiple databases. Cross-corpus studies are conducted to determine the ability of these features to generalize well across different databases. The proposed method is also applied to derive features from facial expressions; a multi-modal fusion overcomes the deficiencies of a speech only approach and further improves the recognition performance.
Besides affecting the acoustic properties of speech, emotions have a strong influence over speech articulation kinematics. A learning approach, which constrains a classifier trained over acoustic descriptors, to also model articulatory data is proposed here. This method requires articulatory information only during the training stage, thus overcoming the challenges inherent to large-scale data collection, while simultaneously exploiting the correlations between articulation kinematics and acoustic descriptors to improve the accuracy of emotion recognition systems.
Identifying context from ambient sounds in a lifelogging scenario requires feature extraction, segmentation and annotation techniques capable of efficiently handling long duration audio recordings; a complete framework for such applications is presented. The performance is evaluated on real world data and accompanied by a prototypical Android-based user interface.
The proposed methods are also assessed in terms of computation and implementation complexity. Software and field programmable gate array based implementations are considered for emotion recognition, while virtual platforms are used to model the complexities of lifelogging. The derived metrics are used to determine the feasibility of these methods for applications requiring real-time capabilities and low power consumption.
ContributorsShah, Mohit (Author) / Spanias, Andreas (Thesis advisor) / Chakrabarti, Chaitali (Thesis advisor) / Berisha, Visar (Committee member) / Turaga, Pavan (Committee member) / Arizona State University (Publisher)
Created2015
Description
Modern machine learning systems leverage data and features from multiple modalities to gain more predictive power. In most scenarios, the modalities are vastly different and the acquired data are heterogeneous in nature. Consequently, building highly effective fusion algorithms is at the core to achieve improved model robustness and inferencing performance. This dissertation focuses on the representation learning approaches as the fusion strategy. Specifically, the objective is to learn the shared latent representation which jointly exploit the structural information encoded in all modalities, such that a straightforward learning model can be adopted to obtain the prediction.
We first consider sensor fusion, a typical multimodal fusion problem critical to building a pervasive computing platform. A systematic fusion technique is described to support both multiple sensors and descriptors for activity recognition. Targeted to learn the optimal combination of kernels, Multiple Kernel Learning (MKL) algorithms have been successfully applied to numerous fusion problems in computer vision etc. Utilizing the MKL formulation, next we describe an auto-context algorithm for learning image context via the fusion with low-level descriptors. Furthermore, a principled fusion algorithm using deep learning to optimize kernel machines is developed. By bridging deep architectures with kernel optimization, this approach leverages the benefits of both paradigms and is applied to a wide variety of fusion problems.
In many real-world applications, the modalities exhibit highly specific data structures, such as time sequences and graphs, and consequently, special design of the learning architecture is needed. In order to improve the temporal modeling for multivariate sequences, we developed two architectures centered around attention models. A novel clinical time series analysis model is proposed for several critical problems in healthcare. Another model coupled with triplet ranking loss as metric learning framework is described to better solve speaker diarization. Compared to state-of-the-art recurrent networks, these attention-based multivariate analysis tools achieve improved performance while having a lower computational complexity. Finally, in order to perform community detection on multilayer graphs, a fusion algorithm is described to derive node embedding from word embedding techniques and also exploit the complementary relational information contained in each layer of the graph.
We first consider sensor fusion, a typical multimodal fusion problem critical to building a pervasive computing platform. A systematic fusion technique is described to support both multiple sensors and descriptors for activity recognition. Targeted to learn the optimal combination of kernels, Multiple Kernel Learning (MKL) algorithms have been successfully applied to numerous fusion problems in computer vision etc. Utilizing the MKL formulation, next we describe an auto-context algorithm for learning image context via the fusion with low-level descriptors. Furthermore, a principled fusion algorithm using deep learning to optimize kernel machines is developed. By bridging deep architectures with kernel optimization, this approach leverages the benefits of both paradigms and is applied to a wide variety of fusion problems.
In many real-world applications, the modalities exhibit highly specific data structures, such as time sequences and graphs, and consequently, special design of the learning architecture is needed. In order to improve the temporal modeling for multivariate sequences, we developed two architectures centered around attention models. A novel clinical time series analysis model is proposed for several critical problems in healthcare. Another model coupled with triplet ranking loss as metric learning framework is described to better solve speaker diarization. Compared to state-of-the-art recurrent networks, these attention-based multivariate analysis tools achieve improved performance while having a lower computational complexity. Finally, in order to perform community detection on multilayer graphs, a fusion algorithm is described to derive node embedding from word embedding techniques and also exploit the complementary relational information contained in each layer of the graph.
ContributorsSong, Huan (Author) / Spanias, Andreas (Thesis advisor) / Thiagarajan, Jayaraman (Committee member) / Berisha, Visar (Committee member) / Tepedelenlioğlu, Cihan (Committee member) / Arizona State University (Publisher)
Created2018
Description
With a growing number of adults with autism spectrum disorder (ASD), more and more research has been conducted on majority male cohorts with ASD from young, adolescence, and some older age. Currently, males make up the majority of individuals diagnosed with ASD, however, recent research states that the gender gap is closing due to more advanced screening and a better understanding of how females with ASD present their symptoms. Little research has been published on the neurocognitive differences that exist between older adults with ASD compared to neurotypical (NT) counterparts, and nothing has specifically addressed older women with ASD. This study utilized neuroimaging and neuropsychological tests to examine differences between diagnosis and sex of four distinct groups: older men with ASD, older women with ASD, older NT men, and older NT women. In each group, hippocampal size (via FreeSurfer) was analyzed for differences as well as correlations with neuropsychological tests. Participants (ASD Female, n = 12; NT Female, n = 14; ASD Male, n = 30; NT Male = 22), were similar according to age, IQ, and education. The results of the study indicated that the ASD Group as a whole performed worse on executive functioning tasks (Wisconsin Card Sorting Test, Trails Making Test) and memory-related tasks (Rey Auditory Verbal Learning Test, Weschler Memory Scale: Visual Reproduction) compared to the NT Group. Interactions of sex by diagnosis approached significance only within the WCST non-perseverative errors, with the women with ASD performing worse than NT women, but no group differences between men. Effect sizes between the female groups (ASD female vs. NT female) showed more than double that of the male groups (ASD male vs. NT male) for all WCST and AVLT measures. Participants with ASD had significantly smaller right hippocampal volumes than NT participants. In addition, all older women showed larger hippocampal volumes when corrected for total intracranial volume (TIV) compared to all older men. Overall, NT Females had significant correlations across all neuropsychological tests and their hippocampal volumes whereas no other group had significant correlations. These results suggest a tighter coupling between hippocampal size and cognition in NT Females than NT Males and both sexes with ASD. This study promotes further understanding of the neuropsychological differences between older men and women, both with and without ASD. Further research is needed on a larger sample of older women with and without ASD.
ContributorsWebb, Christen Len (Author) / Braden, B. Blair (Thesis advisor) / Azuma, Tamiko (Committee member) / Dixon, Maria (Committee member) / Arizona State University (Publisher)
Created2019
Description
Deep neural networks (DNN) have shown tremendous success in various cognitive tasks, such as image classification, speech recognition, etc. However, their usage on resource-constrained edge devices has been limited due to high computation and large memory requirement.
To overcome these challenges, recent works have extensively investigated model compression techniques such as element-wise sparsity, structured sparsity and quantization. While most of these works have applied these compression techniques in isolation, there have been very few studies on application of quantization and structured sparsity together on a DNN model.
This thesis co-optimizes structured sparsity and quantization constraints on DNN models during training. Specifically, it obtains optimal setting of 2-bit weight and 2-bit activation coupled with 4X structured compression by performing combined exploration of quantization and structured compression settings. The optimal DNN model achieves 50X weight memory reduction compared to floating-point uncompressed DNN. This memory saving is significant since applying only structured sparsity constraints achieves 2X memory savings and only quantization constraints achieves 16X memory savings. The algorithm has been validated on both high and low capacity DNNs and on wide-sparse and deep-sparse DNN models. Experiments demonstrated that deep-sparse DNN outperforms shallow-dense DNN with varying level of memory savings depending on DNN precision and sparsity levels. This work further proposed a Pareto-optimal approach to systematically extract optimal DNN models from a huge set of sparse and dense DNN models. The resulting 11 optimal designs were further evaluated by considering overall DNN memory which includes activation memory and weight memory. It was found that there is only a small change in the memory footprint of the optimal designs corresponding to the low sparsity DNNs. However, activation memory cannot be ignored for high sparsity DNNs.
To overcome these challenges, recent works have extensively investigated model compression techniques such as element-wise sparsity, structured sparsity and quantization. While most of these works have applied these compression techniques in isolation, there have been very few studies on application of quantization and structured sparsity together on a DNN model.
This thesis co-optimizes structured sparsity and quantization constraints on DNN models during training. Specifically, it obtains optimal setting of 2-bit weight and 2-bit activation coupled with 4X structured compression by performing combined exploration of quantization and structured compression settings. The optimal DNN model achieves 50X weight memory reduction compared to floating-point uncompressed DNN. This memory saving is significant since applying only structured sparsity constraints achieves 2X memory savings and only quantization constraints achieves 16X memory savings. The algorithm has been validated on both high and low capacity DNNs and on wide-sparse and deep-sparse DNN models. Experiments demonstrated that deep-sparse DNN outperforms shallow-dense DNN with varying level of memory savings depending on DNN precision and sparsity levels. This work further proposed a Pareto-optimal approach to systematically extract optimal DNN models from a huge set of sparse and dense DNN models. The resulting 11 optimal designs were further evaluated by considering overall DNN memory which includes activation memory and weight memory. It was found that there is only a small change in the memory footprint of the optimal designs corresponding to the low sparsity DNNs. However, activation memory cannot be ignored for high sparsity DNNs.
ContributorsSrivastava, Gaurav (Author) / Seo, Jae-Sun (Thesis advisor) / Chakrabarti, Chaitali (Committee member) / Berisha, Visar (Committee member) / Arizona State University (Publisher)
Created2018
Description
With advances in automatic speech recognition, spoken dialogue systems are assuming increasingly social roles. There is a growing need for these systems to be socially responsive, capable of building rapport with users. In human-human interactions, rapport is critical to patient-doctor communication, conflict resolution, educational interactions, and social engagement. Rapport between people promotes successful collaboration, motivation, and task success. Dialogue systems which can build rapport with their user may produce similar effects, personalizing interactions to create better outcomes.
This dissertation focuses on how dialogue systems can build rapport utilizing acoustic-prosodic entrainment. Acoustic-prosodic entrainment occurs when individuals adapt their acoustic-prosodic features of speech, such as tone of voice or loudness, to one another over the course of a conversation. Correlated with liking and task success, a dialogue system which entrains may enhance rapport. Entrainment, however, is very challenging to model. People entrain on different features in many ways and how to design entrainment to build rapport is unclear. The first goal of this dissertation is to explore how acoustic-prosodic entrainment can be modeled to build rapport.
Towards this goal, this work presents a series of studies comparing, evaluating, and iterating on the design of entrainment, motivated and informed by human-human dialogue. These models of entrainment are implemented in the dialogue system of a robotic learning companion. Learning companions are educational agents that engage students socially to increase motivation and facilitate learning. As a learning companion’s ability to be socially responsive increases, so do vital learning outcomes. A second goal of this dissertation is to explore the effects of entrainment on concrete outcomes such as learning in interactions with robotic learning companions.
This dissertation results in contributions both technical and theoretical. Technical contributions include a robust and modular dialogue system capable of producing prosodic entrainment and other socially-responsive behavior. One of the first systems of its kind, the results demonstrate that an entraining, social learning companion can positively build rapport and increase learning. This dissertation provides support for exploring phenomena like entrainment to enhance factors such as rapport and learning and provides a platform with which to explore these phenomena in future work.
This dissertation focuses on how dialogue systems can build rapport utilizing acoustic-prosodic entrainment. Acoustic-prosodic entrainment occurs when individuals adapt their acoustic-prosodic features of speech, such as tone of voice or loudness, to one another over the course of a conversation. Correlated with liking and task success, a dialogue system which entrains may enhance rapport. Entrainment, however, is very challenging to model. People entrain on different features in many ways and how to design entrainment to build rapport is unclear. The first goal of this dissertation is to explore how acoustic-prosodic entrainment can be modeled to build rapport.
Towards this goal, this work presents a series of studies comparing, evaluating, and iterating on the design of entrainment, motivated and informed by human-human dialogue. These models of entrainment are implemented in the dialogue system of a robotic learning companion. Learning companions are educational agents that engage students socially to increase motivation and facilitate learning. As a learning companion’s ability to be socially responsive increases, so do vital learning outcomes. A second goal of this dissertation is to explore the effects of entrainment on concrete outcomes such as learning in interactions with robotic learning companions.
This dissertation results in contributions both technical and theoretical. Technical contributions include a robust and modular dialogue system capable of producing prosodic entrainment and other socially-responsive behavior. One of the first systems of its kind, the results demonstrate that an entraining, social learning companion can positively build rapport and increase learning. This dissertation provides support for exploring phenomena like entrainment to enhance factors such as rapport and learning and provides a platform with which to explore these phenomena in future work.
ContributorsLubold, Nichola Anne (Author) / Walker, Erin (Thesis advisor) / Pon-Barry, Heather (Thesis advisor) / Litman, Diane (Committee member) / VanLehn, Kurt (Committee member) / Berisha, Visar (Committee member) / Arizona State University (Publisher)
Created2018
Description
Glottal fry is a vocal register characterized by low frequency and increased signal perturbation, and is perceptually identified by its popping, creaky quality. Recently, the use of the glottal fry vocal register has received growing awareness and attention in popular culture and media in the United States. The creaky quality that was originally associated with vocal pathologies is indeed becoming “trendy,” particularly among young women across the United States. But while existing studies have defined, quantified, and attempted to explain the use of glottal fry in conversational speech, there is currently no explanation for the increasing prevalence of the use of glottal fry amongst American women. This thesis, however, proposes that conversational entrainment—a communication phenomenon which describes the propensity to modify one’s behavior to align more closely with one’s communication partner—may provide a theoretical framework to explain the growing trend in the use of glottal fry amongst college-aged women in the United States. Female participants (n = 30) between the ages of 18 and 29 years (M = 20.6, SD = 2.95) had conversations with two conversation partners, one who used quantifiably more glottal fry than the other. The study utilized perceptual and quantifiable acoustic information to address the following key question: Does the amount of habitual glottal fry in a conversational partner influence one’s use of glottal fry in their own speech? Results yielded the following two findings: (1) according to perceptual annotations, the participants used a greater amount of glottal fry when speaking with the Fry conversation partner than with the Non Fry partner, (2) statistically significant differences were found in the acoustics of the participants’ vocal qualities based on conversation partner. While the current study demonstrates that young women are indeed speaking in glottal fry in everyday conversations, and that its use can be attributed in part to conversational entrainment, we still lack a clear explanation of the deeper motivations for women to speak in a lower vocal register. The current study opens avenues for continued analysis of the sociolinguistic functions of the glottal fry register.
ContributorsDelfino, Christine R (Author) / Liss, Julie M (Thesis advisor) / Borrie, Stephanie A (Thesis advisor) / Azuma, Tamiko (Committee member) / Berisha, Visar (Committee member) / Arizona State University (Publisher)
Created2015
Description
Many neurological disorders, especially those that result in dementia, impact speech and language production. A number of studies have shown that there exist subtle changes in linguistic complexity in these individuals that precede disease onset. However, these studies are conducted on controlled speech samples from a specific task. This thesis explores the possibility of using natural language processing in order to detect declining linguistic complexity from more natural discourse. We use existing data from public figures suspected (or at risk) of suffering from cognitive-linguistic decline, downloaded from the Internet, to detect changes in linguistic complexity. In particular, we focus on two case studies. The first case study analyzes President Ronald Reagan’s transcribed spontaneous speech samples during his presidency. President Reagan was diagnosed with Alzheimer’s disease in 1994, however my results showed declining linguistic complexity during the span of the 8 years he was in office. President George Herbert Walker Bush, who has no known diagnosis of Alzheimer’s disease, shows no decline in the same measures. In the second case study, we analyze transcribed spontaneous speech samples from the news conferences of 10 current NFL players and 18 non-player personnel since 2007. The non-player personnel have never played professional football. Longitudinal analysis of linguistic complexity showed contrasting patterns in the two groups. The majority (6 of 10) of current players showed decline in at least one measure of linguistic complexity over time. In contrast, the majority (11 out of 18) of non-player personnel showed an increase in at least one linguistic complexity measure.
ContributorsWang, Shuai (Author) / Berisha, Visar (Thesis advisor) / LaCross, Amy (Committee member) / Tong, Hanghang (Committee member) / Arizona State University (Publisher)
Created2016
Description
The purpose of this study was to identify acoustic markers that correlate with accurate and inaccurate /r/ production in children ages 5-8 using signal processing. In addition, the researcher aimed to identify predictive acoustic markers that relate to changes in /r/ accuracy. A total of 35 children (23 accurate, 12 inaccurate, 8 longitudinal) were recorded. Computerized stimuli were presented on a PC laptop computer and the children were asked to do five tasks to elicit spontaneous and imitated /r/ production in all positions. Files were edited and analyzed using a filter bank approach centered at 40 frequencies based on the Mel-scale. T-tests were used to compare spectral energy of tokens between accurate and inaccurate groups and additional t-tests were used to compare duration of accurate and inaccurate files. Results included significant differences between the accurate and inaccurate productions of /r/, notable differences in the 24-26 mel bin range, and longer duration of inaccurate /r/ than accurate. Signal processing successfully identified acoustic features of accurate and inaccurate production of /r/ and candidate predictive markers that may be associated with acquisition of /r/.
ContributorsBecvar, Brittany Patricia (Author) / Azuma, Tamiko (Thesis advisor) / Weinhold, Juliet (Committee member) / Berisha, Visar (Committee member) / Arizona State University (Publisher)
Created2017
Description
In this pilot study, the purpose was to determine if certain language interventions could help bilingual children reduce maze use and improve their story retell abilities. We used language intervention, Story Champs, and its Spanish version, Puente de Cuentos to help bilingual children improve their story retell abilities. We conducted the intervention over the course of three days in both Spanish and English. The children participated in three stories in each language each day. They also received a narrative measure before and after the intervention to measure gains in story ability and to measure maze use. Results of the study indicated that there were no statistically-significant differences in the children's story retell abilities or maze use before and after the intervention. Nevertheless, we are encouraged by our results for future further study because of some improvements the children made.
ContributorsWhiteley, Aaron Kyle (Author) / Restrepo, Maria (Thesis director) / Dixon, Maria (Committee member) / Department of Speech and Hearing Science (Contributor) / School of International Letters and Cultures (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05