1,053 Hits in 4.6 sec

Analyzing analytical methods: The case of phonology in neural models of spoken language [article]

Grzegorz Chrupała, Bertrand Higy, Afra Alishahi
2020 arXiv   pre-print
We use two commonly applied analytical techniques, diagnostic classifiers and representational similarity analysis, to quantify to what extent neural activation patterns encode phonemes and phoneme sequences  ...  local-scope diagnostic methods.  ...  Others predominantly use diagnostic classifiers for phoneme and grapheme classification from neural representations of speech.  ... 
arXiv:2004.07070v2 fatcat:a43fku5hrzfcfa2mqewbimrwh4

An Automatic Pronunciation Error Detection and Correction Mechanism in English Teaching Based on an Improved Random Forest Model

Yuhua Dai, Wei Liu
2022 Journal of Electrical and Computer Engineering  
Mel cepstral coefficient (MFCC) is used for feature extraction, and principal component analysis (PCA) is used for dimensionality reduction of feature data.  ...  The experimental structure demonstrates that by using a combination classification framework based on MFCC, PCA, and RF, the learner's pronunciation difficulty may be resolved.  ...  Speech Signal Processing Method. ere are many categories of speech recognition systems based on different classifications, such as isolated word or continuous speech recognition, speaker-specific and speaker-nonspecific  ... 
doi:10.1155/2022/6011993 fatcat:taxj3puvk5hxdfmlwkmof3obde

"The Diagnostic Evaluation of Switchboard-corpus Automatic Speech Recognition Systems"

Madhav Singh Solanki
2021 International Journal of Innovative Research in Computer Science & Technology  
The decision trees show that correct categorization of phonetic segments and characteristics is one of the most constant variables linked with better recognition performance.  ...  recognition systems.  ...  projection with a smaller dimension. a certain speaker Acoustic voice recognition is used by the majority of speech recognition systems [8] .  ... 
doi:10.55524/ijircst.2021.9.6.5 fatcat:btg6zz6tqfbknacbfb3wkkhmu4

Automatic detection of articulation disorders in children with cleft lip and palate

Andreas Maier, Florian Hönig, Tobias Bocklet, Elmar Nöth, Florian Stelzle, Emeka Nkenke, Maria Schuster
2009 Journal of the Acoustical Society of America  
To create a reference for an automatic system, speech data of 58 children with CLP were assessed perceptually by experienced speech therapists for characteristic phonetic disorders at the phoneme level  ...  Speech of children with cleft lip and palate ͑CLP͒ is sometimes still disordered even after adequate surgical and nonsurgical therapies.  ...  Ulrike Wohlleben, Andrea Schädel, and Dorothee Großmann for the expert's annotation of the data.  ... 
doi:10.1121/1.3216913 pmid:19894838 fatcat:ufz4hkqdybehzg2s452gyud2dy

Phone Duration Modeling for Speaker Age Estimation in Children [article]

Prashanth Gurunath Shivakumar, Somer Bishop, Catherine Lord, Shrikanth Narayanan
2021 arXiv   pre-print
Statistical functionals are computed from phone duration distributions for each phoneme which are in turn used to train regression models to predict speaker age.  ...  Experimental results suggest phone durations contain important development-related information of children. Phonemes contributing most to estimation of children speaker age are analyzed and presented.  ...  Whereas, the decision tree based AdaBoost model has the benefit of providing feature importance which helps us analyze the contributions of phonemes and their discriminative power towards age estimation  ... 
arXiv:2109.01568v1 fatcat:4tskn743xngurbreotj2n2nvla

Learning diagnostic models using speech and language measures

Bart Peintner, William Jarrold, Dimitra Vergyri, Colleen Richey, Maria Luisa Gorno Tempini, Jennifer Ogar
2008 2008 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society  
We describe results that show the effectiveness of machine learning in the automatic diagnosis of certain neurodegenerative diseases, several of which alter speech and language production.  ...  From this data, we extracted features of the audio signal and the words the patient used, which were obtained using our automated transcription technologies.  ...  From this data, we extracted features of the audio signal and the words the patient used, which were obtained using our automatic speech recognition (ASR) algorithms, our phoneme duration measurement tools  ... 
doi:10.1109/iembs.2008.4650249 pmid:19163752 fatcat:we3guradxzhddjgowadzw6opqe

Discrete representations in neural models of spoken language [article]

Bertrand Higy, Lieke Gelderloos, Afra Alishahi, Grzegorz Chrupała
2021 arXiv   pre-print
While we can attribute them to the properties of the different metrics in most cases, one point of concern remains: the use of minimal pairs of phoneme triples as stimuli disadvantages larger discrete  ...  We compare the merits of four commonly used metrics in the context of weakly supervised models of spoken language.  ...  We would also like to thank multiple anonymous reviewers for their useful comments which helped us improve this paper.  ... 
arXiv:2105.05582v2 fatcat:oylbownbbzbrracnzfe7eodcbi

Detecting Mispronunciations of L2 Learners and Providing Corrective Feedback Using Knowledge-Guided and Data-Driven Decision Trees

Wei Li, Kehuang Li, Sabato Marco Siniscalchi, Nancy F. Chen, Chin-Hui Lee
2016 Interspeech 2016  
We propose a novel decision tree based framework to detect phonetic mispronunciations produced by L2 learners caused by using inaccurate speech attributes, such as manner and place of articulation.  ...  and vowels are produced using related articulators, can be provided to the L2 learners; and (3) by building the phone-dependent decision tree, the relative importance of the speech attribute features  ...  Speech Attribute Classification System Setup The input feature (see Figure 1 ) is a window of 11 speech frames, each includes a 39-dim MFCC+Δ+ΔΔ vector.  ... 
doi:10.21437/interspeech.2016-517 dblp:conf/interspeech/LiLSCL16 fatcat:feca2njtazcgdkyapusxfillqq

Acoustic-Based Articulatory Phenotypes of Amyotrophic Lateral Sclerosis and Parkinson's Disease: Towards an Interpretable, Hypothesis-Driven Framework of Motor Control

Hannah P. Rowe, Sarah E. Gutz, Marc F. Maffei, Jordan R. Green
2020 Interspeech 2020  
The use of interpretable, hypothesis-driven features has the potential to inform impairment-based automatic speech recognition (ASR) models and improve classification algorithms for disorders with divergent  ...  With additional research, articulatory phenotypes characterized using this framework may lead to advancements in ASR for dysarthric speech and diagnostic accuracy at different disease stages for individuals  ...  Data classification was also performed using ensemble decision trees with leave-one-out cross-validation (subsampling).  ... 
doi:10.21437/interspeech.2020-1459 dblp:conf/interspeech/RoweGMG20 fatcat:r4nywsjzrzcndgrce7pgajrxhi

An Arabic Mispronunciation Detection System Based on the Frequency of Mistakes for Asian Speakers

Faria Nazir, Muhammad Nadeem Majeed, Mustansar Ali Ghazanfar, Muazzam Maqsood
2021 Mehran University Research Journal of Engineering and Technology  
We use the Support Vector Machine (SVM) classifier and test the results on the Arabic dataset (28 Phonemes). The performance of our proposed method is evaluated by using accuracy.  ...  We consider mispronunciation detection as a classification problem, traditionally for this purpose, a separate classifier is trained for each phoneme mistake that requires a lot of memory and time.  ...  Tabassam Nawaz, Chairman, and Associate Professor of the Software Engineering Department, University of Engineering and Technology, Taxila, Pakistan for providing the cooperative environment and infrastructure  ... 
doi:10.22581/muet1982.2102.03 fatcat:pffogp7thna4fh3xl275ptvtwa

Language in Amnestic Mild Cognitive Impairment and Dementia of Alzheimer's Type: Quantitatively or Qualitatively Different?

Regina Jokel, Bruna Seixas Lima, Alita Fernandez, Kelly J. Murphy
2019 Dementia and Geriatric Cognitive Disorders Extra  
diagnostic entity or represent two distinct cognitive disorders.  ...  Conclusion: Language tests provide an important contribution to the diagnostic process in their capacity to identify language impairments at an early stage.  ...  In addition, information pertaining to motor speech (such as presence of apraxia or dysarthria) and the use of correct syntax in spontaneous speech was extracted from speech-language pathology reports.  ... 
doi:10.1159/000496824 fatcat:f3a5p3ntezgqpb4qc2xekef5va

Description and distribution of the subtypes of chronic schizophrenia based on Leonhard's classification

T A Ban, W Guy, W H Wilson
1984 Psychiatric developments  
With the recognition of neuroleptic response heterogeneity, and of the hazards of long-term neuroleptic administration, interest in subtype classification such as that of Leonhard has grown.  ...  Prior to the introduction of neuroleptics a lack of interest in complex classifications of the chronic schizophrenias was due to the lack of effective treatment.  ...  With the recognition of neuroleptic response heterogeneity, and of the hazards of long-term neuroleptic administration, interest in subtype classification such as that of Leonhard has grown.  ... 
pmid:6514708 fatcat:ftjej2jtgzbrjbavtdnoddne7i

A Novel Method for Feature Extraction in Vocal Fold Pathology Diagnosis [chapter]

Vahid Majidnezhad, Igor Kheidorov
2013 Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering  
While the second stage implies a choice of a variety of machine learning methods, the first stage plays a critical role in performance of the classification system.  ...  Support vector machine is used as a classifier for evaluating the performance of our proposed method. The results show the priority of the proposed method in comparison with other methods.  ...  This work was supported by the speech laboratory of NASB in Belarus.  ... 
doi:10.1007/978-3-642-37893-5_11 fatcat:n3jis4zxcraardonavsjcdugou


Klaus Zechner, Isaac I. Bejar, Ramin Hemat
2007 ETS Research Report Series  
In this investigation, we evaluated the feasibility of using an off-the-shelf speech-recognition system for scoring speaking prompts from the LanguEdge field test of 2002.  ...  We then adapted a speech engine to the language backgrounds and proficiency ranges of the speakers and developed a classification and regression tree (CART) for each of five prompts based on features computed  ...  The tree was then validated on a portion of the sample that had not been used in estimating the classification trees.  ... 
doi:10.1002/j.2333-8504.2007.tb02044.x fatcat:tsnlv5zc3bhr5bxf3ghz5uitrm


1987 Phonetica: International Journal of Phonetic Science  
One of the more recent classifications of speech sounds by Lade-foged [1975] relies heavily on Pike.  ...  very useful in a description of phones and phonemes in terms of mono-and polysegmentality.  ... 
doi:10.1159/000261779 fatcat:63dobj5fmnc3jfnmw5hwui3od4
« Previous Showing results 1 — 15 out of 1,053 results