Accurate Speech Research Articles

Use of Voice Recognition Technology and Simple Visual Warning Systems on Forklifts to Detect Presence Forklift to Improve Communication and Safety of People Around and the Work Environment

This research explores the integration of voice recognition tech and visual warnings on forklifts to enhance safety for bystanders and workplace environments. It employs voice recognition for driver instructions and a straightforward visual warning system using colored lights. Prototype testing displayed promising outcomes with accurate speech recognition and clear visual cues. This technology holds promise in enhancing communication and safety by enabling effective warnings from forklift drivers and facilitating appropriate precautions for bystanders. The study contributes to advancing safety and well-being in forklift use within workplaces.

Preliminary Speech Perception Performance Profiles of School-Age Children With Childhood Apraxia of Speech, Speech Sound Disorder, and Typical Development.

Limited research exists assessing speech perception in school-age children with speech sound disorder (SSD) and childhood apraxia of speech (CAS); despite early evidence that speech perception may lead to error-prone motor planning/programming. In this study, we examine speech perception performance in school-age children with and without speech production deficits. Speech perception was assessed using the Wide Range Acoustic Accuracy Scale to determine the just-noticeable difference in discrimination for three consonant-vowel syllable contrasts (/bɑ/-/wɑ/, /dɑ/-/gɑ/, /ɹɑ/-/wɑ/), each varying along a single acoustic parameter for seven children with CAS with rhotic errors, seven children with SSD with rhotic errors, and seven typically developing (TD) children. Findings revealed statistically significant mean differences between perceptual performance of children with CAS when compared to TD children for discrimination of /ɹɑ/-/wɑ/ contrasts. Large effect sizes were also observed for comparisons of /ɹɑ/-/wɑ/ contrasts between children with CAS, SSD, and TD peers. Additionally, large effect sizes were observed for /dɑ/-/gɑ/ contrasts between children with CAS and SSD and TD children despite nonsignificant mean differences in group performance. Overall, mean outcome scores suggest that school-age children with CAS and persistent rhotic errors demonstrated less accurate speech perception skills relative to TD children for the /ɹɑ/-/wɑ/ contrasts. However, the relatively small sample sizes per group limit the extent to which these findings may be generalized to the broader population.

The effects of using L1 Chinese or L2 English in planning on speaking performance among high- and low-proficient EFL learners

Speaking constitutes one of the main goals of learning a second language (L2). Despite the increasing attention on the role of planning and language transfer in L2 learning, the combined effect of using different languages and pre-task planning on language production remains unclear. This study investigated whether the use of different languages in planning affects speaking performance and whether the effect differs by language proficiency. A total of 84 students in Chinese universities learning English as a foreign language participated in several speaking tasks after planning using their first language (L1) Chinese or L2 English. Findings showed that using L1 in planning results in significantly higher syntactic complexity, accuracy, and fluency in speaking performance than using L2 in planning, while the difference in lexical diversity were not statistically significant. Further analysis shows that for speech accuracy, the facilitative effect of L1 was stronger among low-proficient than high-proficient learners. Findings from this study support the use of L2 learners’ entire linguistic repertoire in speaking activities and provides implications on speech production theories as well as translanguaging pedagogies.

Development of efficient techniques for ASR System for Speech Detection and Recognization system using Gaussian Mixture Model- Universal Background Model

Some practical uses of ASR have been implemented, including the transcription of meetings and the usage of smart speakers. It is the process by which speech waves are transformed into text that allows computers to interpret and act upon human speech. Scalable strategies for developing ASR systems in languages where no voice transcriptions or pronunciation dictionaries exist are the primary focus of this work. We first show that the necessity for voice transcription into the target language can be greatly reduced through cross-lingual acoustic model transfer when phonemic pronunciation lexicons exist in the new language. Afterwards, we investigate three approaches to dealing with languages that lack a pronunciation lexicon. Secondly, we have a look at the efficiency of graphemic acoustic model transfer, which makes it easy to build pronunciation dictionaries. Thesis problems can be solved, in part, by investigating optimization strategies for training on huge corpora (such as GA+HMM and DE+HMM). In the training phase of acoustic modelling, the suggested method is applied to traditional methods. Read speech and HMI voice experiments indicated that while each data augmentation strategy alone did not always increase recognition performance, using all three techniques together did. Power normalised cepstral coefficient (PNCC) features are tweaked somewhat in this work to enhance verification accuracy. To increase speaker verification accuracy, we suggest employing multiple “Gaussian Mixture Model-Universal Background Model (GMM-UBM) and SVM classifiers”. Importantly, pitch shift data augmentation and multi-task training reduced bias by more than 18% absolute compared to the baseline system for read speech, and applying all three data augmentation techniques during fine tuning reduced bias by more than 7% for HMI speech, while increasing recognition accuracy of both native and non-native Dutch speech.

“Like setting down on a kag of powder and touching it off just to see where you’ll go to”: Reflections on Forty Years of Teaching Huckleberry Finn

Abstract A pedagogical essay reflecting on forty years of teaching Huckleberry Finn at both the high school and college level, recounting the author’s first experience teaching the novel, the expansion of the canon and the way that puts the novel in a broader context, and the importance of performing the dialect speech of various characters, modeling for students the way to read the novel for themselves. Even in perilous times for teaching controversial texts, the author suggests ways to teach Mark Twain’s masterpiece sensitively and effectively.

A corpus study on the relationship between focus and accent in natural speech in Mandarin Chinese

This paper aims to explore the relationship between focus and accent, namely, whether the focus of a sentence possesses some acoustic features of accent. Previous studies suggest that focused words depict different patterns of acoustic prominence embedded in accent. According to previous evidence found in English, pitch and duration are two crucial factors contributing to differentiating focused and unfocused words. Therefore, pitch and duration are the main two factor considered in this study. Studies in aspect of the relationship between focus and accent and their realization in the phonetic level in natural speech are insufficient. Therefore, this study aims to provide support for the connection between focus and accent in natural speech in Mandarin. Based on 24 sentences from three native Mandarin speakers selected from a Chinese corpus, THCHS-30, published by Tsinghua University, this study extracts the narrow-focused words, broad-focused words, and the corresponding unfocused-words as the main subjects and analyses their mutual relationships. Results support the symmetric relationship between focus and accent. Focused words show phonetic features of accent and words with no obvious phonetic prominence of accent are unlikely to be considered as focus. Focus triggers a decrease in F0 values and an increase in duration of focused words. Linear mixed effects tests in R suggest that a major effect of narrow and broad focus in causing F0 and duration changes has not been deducted, but narrow focus conditions can result in duration proportion increase of narrow focused words compared to broad focused ones.

VOCABULARY OF THE MIDDLE BUG AREA DIALECTS IN THE DIALECT DICTIONARY

The article presents a fragment of the “Dictionary of the dialects of the Middle Bug area”. The object of the study is agricultural vocabulary documented in the dialects of the Middle Bug area. The subject of the study is the organization of this lexical layer in the dialect dictionary. The sources of the factual material are audio recordings of vernacular speech with a total duration of 205 hours 54 min. 13 s., made in 74 settlements of the Middle Bug area during the period of 2014–2016. The article uses general scientific methods of research: analysis, synthesis and observation, which made it possible to identify and systematize the dialect material; the descriptive method was used in the analysis of vernacular speech. Special linguistic research methods: questionnaires and interviews contributed to the documenting of dialect speech, contextual and interpretive analysis allowed to identify transformations of the semantic scope of lexemes functioning in the speech of speakers of the MiddleBug dialects, linguogeographical method made it possible to illustrate the areology of the documented names. The article describes the structure of a dictionary entry and presents a fragment of the “Dictionary of the dialects of the Middle Bug area”. The register lexemes are marked with letter abbreviations of the names of the examined settlements and with the references to the “Etymological Dictionary of the Ukrainian Language” (1982–2012), the “Eleven-volume Dictionary of the Ukrainian Language” (1970–1980), the “Dictionary of the Ukrainian Language edited by B. D. Hrinchenko” (1907), and the “Dictionary of Ukrainian Dialects of Odesa Region” edited by O. I. Bondar (2011) if the respective word is documented as a dialectism in the above lexicographic sources. In order to reveal the semantics of a register word and describe in more detail the specifics of the Middle Bug area dialects, sentences and microtexts of dialect speech with the obligatory indication of the settlement where they were recorded have been added. The peculiarities of pronunciation are reflected and illustrative material is presented using the general phonetic transcription for recording Ukrainian literary and dialect speech, which is proposed in the first volume of the “Atlas of the Ukrainian Language”. The analyzed lexical material emphasizes the originality of the Middle Bug area dialects, as evidenced by pronounced lexical, phonetic and, less often, accentual and grammatical variants of dialect names.

Research on the application of machine learning models in speech recognition in noisy environments

In a world increasingly reliant on verbal communication with machines, this paper undertakes a rigorous exploration of the application of machine learning models in speech recognition within noisy environments. An important focus of this research lies in demonstrating the viability and robustness of these models when challenged by real-world noise interference. In particular, the utilization of deep learning models such as Convolutional Neural Networks (CNNs) and Long Short-Term Memory Networks (LSTMs) is thoroughly examined. Using concrete data and case studies, this paper uncovers a promising trajectory in the development and implementation of noise-robust speech recognition systems. In so doing, it lays a foundation for future research to enhance the practicality and effectiveness of machine learning models in this area. The ultimate goal is to ensure more reliable and accurate speech recognition that can withstand various noise conditions, thus expanding the potential for speech recognition applications in diverse and adverse environments.

Contrastive stress in persons with Parkinson's disease who speak Mandarin: Task effect in production and preserved perception

BackgroundSpeech in persons with Parkinson's disease (PD) is characterized by impaired prosody (e.g., monotone, abnormal rate, reduced loudness). Most studies on prosodic abnormalities in PD have been obtained from individuals who speak non-tone languages, where prosodic contrasts do not systematically contribute to lexical meanings. In a tone language such as Mandarin, pitch not only carries affective information but also serves to distinguish lexical meanings. It is not known how well persons with PD, who speak a tone language, convey contrastive stress (specific intonational cues signaling topic and theme) for discourse purposes in production, or how well they perceive these contrasts. MethodExperiment 1 investigated production of contrastive stress by persons with PD who speak Mandarin using two different speech tasks, Elicitation and Repetition. PD participants and healthy controls (HC) produced short sentences with focus in different positions during the two task conditions. As an indirect measurement of the quality of the participants’ production of contrastive stress, healthy listeners served as raters to identify focus positions in the sentences and provide goodness ratings to each produced contrastive stress. Experiment 2 examined perceptual ability, measuring PD participants' identification through listening of contrastive stress on utterances produced by a healthy speaker. ResultsFor the Production Study (Experiment 1), the results revealed significantly poorer performance in the PD than the HC group in Elicitation and Repetition. Consistent with previous studies, a task effect was found; study participants demonstrated better performance in Repetition than in Elicitation. Results for the examination of perceptual ability in Experiment 2 revealed that PD and HC participants were equally successful in perceiving contrastive stress in Mandarin utterances produced by a healthy speaker. DiscussionThis study extended previous literature by measuring production and perception of contrastive stress in persons with PD who speak a tone language. Contrastive stress was detected with decreased accuracy in speech produced by persons with PD compared to healthy controls. However, performance was relatively preserved in a repetition condition compared to an elicitation condition. In contrast to the production results, speakers with PD were as successful as HC in perceiving sentential focus, consistent with previous research reporting a discrepancy between production and perception in persons with PD.

A - 87 A Neuropsychological Case Study of Primary Progressive Apraxia of Speech Versus Agrammatic Primary Progressive Aphasia.

Primary progressive apraxia of speech (PPAOS) is a lesser-known neurodegenerative disorder characterized by progressive speech production loss, affecting word articulation. Apraxic speech commonly occurs in primary progressive aphasias (PPAs) and as a result, is often misdiagnosed as agrammatic PPA (aPPA). The current case highlights the importance of neuropsychological evaluations and behavioral observations in delineating PPAOS from aPPA. The patient is a 71-year-old right-handed White female who presented with a 2-year history of language difficulties beginning with a lisp and progressing to an inability to produce words. MRI of the brain showed moderate white matter disease. Her neurocognitive profile was notable for primary deficits in expressive language, including oral/written production, verbal fluency, and repetition. Milder deficits were observed in aspects of executive functioning. Observed spontaneous speech was notable for impaired and effortful articulation with syllabically segmented prosodic speech patterns, some distortions and prosody alterations, and slow rate. All other cognitive domains were intact. Patients with PPAOS and aPPA typically score in the impaired range on tests that depend on rate and accuracy of speech. Apraxia was her most salient feature reported and observed on testing. We did not find learning and memory difficulties, which are more common in aPPA. Moreover, she had trial-and-error articulatory attempts, groping, and distorted substitutions, which are more commonly seen in PPAOS. Her speech was also devoid of grammatical errors which is another distinguishing feature. While subtle and confounded by her alcohol use and orthopedic injuries, her profile altogether appeared most consistent with PPAOS.

Children's listening effort for L2-accented speech

Speech perception is a complex task that can become challenging in certain conditions, such as listening to speech in unfamiliar accents. Using task-evoked pupillary response (TEPR) and dual-task procedures, prior research has shown that adults recruit additional cognitive resources when processing fully intelligible L2-accented speech compared to speech in their own L1 accent (McLaughlin and Van Engen, 2020; Brown et al., 2020). In the current study, we used pupillometry to investigate this effect in children (n = 35, ages 5;0 – 8;11) and ask whether prosodic variation can explain any differences in listening effort. Using Growth Curve Analysis, we found that unlike adults, children do not appear to recruit extra cognitive resources to process intelligible L2-accented speech. Like adults, however, pupillary response declined more rapidly for L2-accented speech across the experiment than for L1-accented speech, indicating either rapid adaptation or fatigue. Additionally, we identified three prosodic measures (relative word duration, pitch stability, and pitch range) that were related to speech adaptation/fatigue for both accent conditions, but in opposite directions. Although children do not appear to upregulate listening effort for L2-accented speech like adults, they appear to use prosodic cues differently for L1- versus L2- accented speech as they adapt to speech.

Identifying linguistic contributions to variable speech perception in Spanish/English bilingual children and adults using the ascending target-to-masker ratio method

When people encounter variable speech, like an unfamiliar accent, they map new productions to existing word forms. This skill appears to develop through adolescence for monolingual children, with school-age children perceiving unfamiliar accented speech less accurately than adults. For monolingual children, vocabulary size and phonemic awareness predict performance. Less is known about which factors account for individual differences in variable speech perception in multilingual school-age children. The goals of this study were to (a) assess group differences in variable speech perception in Spanish/English bilingual and English monolingual children and adults, (b) validate the ascending target-to-masker (TMR) method for assessing speech reception thresholds (SRTs) for foreign accented speech, and (c) identify linguistic skills that account for individual differences in performance. Participants completed speech perception and language testing. SRTs in speech-shaped noise were measured for Midland-, Spanish-, and Korean-accented English; these accents ranged from familiar to unfamiliar across participants. In each accent condition, one list of Bamford-Kowal-Bench test sentences was presented, with each sentence presented at multiple TMRs. Preliminary data suggest differences in how bilingual and monolingual groups perceive variable speech in noise. While linguistic skills accounted for individual differences in performance, different patterns emerged in children and adults.

Machine recognition of non-native speech: Task-specific language models versus large language models

Automatic speech recognition (ASR) has offered a reliable foundation for measurement of young children’s reading skills and of second-language (L2) speaking skills. This is because well-fit task-specific language models (LMs) enable recognition that supports accurate scoring of pronunciation, fluency, vocabulary, usage, and grammar (Bernstein and Cheng, 2023). ASR works well in these measurement tasks because measurement of word production, disfluencies, and pronunciation errors is not very sensitive to moderate differences in word-error-rate (WER) accuracy, and because speech-interactive tasks appropriate for reading instruction or L2 assessment elicit relatively predictable responses, for which task-specific low-perplexity ASR systems achieve sufficiently accurate speech recognition (Cheng and Townshend, 2003). In the work reported here, we compared the accuracy of two English ASR systems on a set of 718 extended spontaneous speech recordings from 77 adult non-native speakers of English speaking from six countries under uncontrolled recording conditions. A Kaldi-based ASR system with well-fit task-specific LMs achieved WER 17%, while USM, a general-purpose mSLAM recognizer with an RNN-T decoder, achieved 11% WER, which is a 34% relative improvement. The mSLAM + RNN-T technology will be briefly described and an analysis of results in three different open-response interactive speaking tasks will be presented.

Efficient measurement of spoken English skills: Design and validation of a 6-minute assessment

Efficient measurement of spoken language skills can improve educational and commercial services. Traditional spoken language assessments often require 30–90 minutes. We report the development and validation of a prototype mobile-browser test of spoken English, “APro”, that participants complete in 6–8 minutes. The APro test presents a stratified random draw of 20 interactive tasks (from pool of 377) to participants whose spoken responses are automatically scored, scaled, and reported. APro aggregates skill-specific measures from disparate task types such as open descriptions and story retellings. Each skill-specific score was independently validated, then combined into an overall spoken English score. APro scoring was trained on 30,000 responses from 354 non-native pilot-test participants who connected remotely from their nine home countries on five continents. The final APro instrument was validated on an independent sample of 79 participants resident in seven countries. In the validation set, human and machine scoring of same-test APro administrations correlated r = 0.98, suggesting accurate speech processing. APro scores predicted much of the score variance in traditional concurrent assessments from Oxford, r = 0.86, and Pearson, r = 0.92. Finally, we present detailed results on two new interactive tasks and share changes to the content, procedure, and scoring that yield improved accuracy at shorter duration. [Work supported in part by Google LLC.]

EXCELLENT ELEMENTS OF PRESENTATION IN BUSINESS ARENA

Presentation is very necessary part of organisation. One should be very careful to make presentation. Always keeps some points to make effective presentation like data collection, accuracy in speech, chronological order of content, use of Multimedia tools. Through an effective presentation it can be convey an aim and objectives. The presentation ideas should be well adapted to your audience. Relate your presentation message/idea to the interests of the audience. It has been creating overall image of organisation.

Analysis of Formant Frequencies and Vowel Articulation in the Spoken Standard Nigerian English of Undergraduate Students

The study compares the first two formants' frequencies in the speech of male and female undergraduate speakers of Standard Nigerian English at the University of Lagos (UNILAG). The study examined eight monophthongs produced in the CVC context (hVd) by 40 males and 40 females, and analysed the vowel quality and phonetic vowel space used by the two gender groups using the Linear Predictive Coding and spectrogram techniques. The findings reveal that there are differences in vowel realisation and quality between male and female speakers, which affect their accents. Specifically, the male undergraduates exhibit compression in the phonetic vowel space while their female counterparts show expansion. This study contributes to the understanding of the vowel space in the speech of male and female speakers, and how these differences may influence speech accent.

Non-native listeners perceiving foreign accentedness of English speech: new evidence from Chinese and Spanish listeners

As an international language, English is an important tool for people of different native languages to communicate with each other. Many empirical studies, proving that foreign accents are prevalent in L2 learners’ oral English, involved native speakers as raters of foreign accents. This study compared the rating patterns of listeners from Chinese, Spanish and English language backgrounds with different English proficiencies for foreign accents in English speech and investigated the acoustical cues used by these listeners. Results showed that Chinese listeners, sharing language backgrounds with talkers, tended to be more tolerant for the Chinese accent in English speech. Spanish listeners had a stricter criterion for mild Chinese accents, proving that the influence of listeners’ L1 on their perception of foreign accents can be different for different degrees of accents. The effects of listeners’ L2 proficiency was found for Chinese listeners’ perception of accents and the number of acoustic predictors used by non-native English listeners. The study provides more evidence about the mechanism of native and non-native listeners’ perception of foreign accent and offers implications for the recruitment of raters for the assessment of oral English.

Speaker Profiling Based on the Short-Term Acoustic Features of Vowels

Speech samples can provide valuable information regarding speaker characteristics, including their social backgrounds. Accent variations with speaker backgrounds reflect corresponding acoustic features of speech, and these acoustic variations can be analyzed to assist in tracking down criminals from speech samples available as forensic evidence. Speech accent identification has recently received significant consideration in the speech forensics research community. However, most works have utilized long-term temporal modelling of acoustic features for accent classification and disregarded the stationary acoustic characteristics of particular phoneme articulations. This paper analyzes short-term acoustic features extracted from a central time window of English vowel speech segments for accent discrimination. Various feature computation techniques have been compared for the accent classification task. It has been found that using spectral features as an input gives better performance than using cepstral features, with the lower filters contributing more significantly to the classification task. Moreover, detailed analysis has been presented for time window durations and frequency bin resolution to compute short-term spectral features concerning accent discrimination. Using longer time durations generally requires higher frequency resolution to optimize classification performance. These results are significant, as they show the benefits of using spectral features for speaker profiling despite the popularity of cepstral features for other speech-related tasks.

The Influence of Stimulus Composition and Scoring Method on Objective Listener Assessments of Tracheoesophageal Speech Accuracy

Introduction: This study investigated the influence of stimulus composition for three speech intelligibility word lists and two scoring methods on the speech accuracy judgments of five tracheoesophageal (TE) speakers. This was achieved through phonemic comparisons across TE speakers’ productions of stimuli from the three intelligibility word lists, including the (1) Consonant Rhyme Test, (2) Northwestern Intelligibility Test, and (3) the Weiss and Basili list. Methodology: Fifteen normal-hearing young adults served as listeners; all listeners were trained in phonetic transcription (IPA), but none had previous exposure to any mode of postlaryngectomy alaryngeal speech. Speaker stimuli were presented to all listeners through headphones, and all stimuli were transcribed phonetically using an open-set response paradigm. Data were analyzed for individual speakers by stimulus list. Phonemic scoring was compared to a whole-word scoring method, and the types of errors observed were quantified by word list. Results: Individual speaker variability was noted, and its effect on the assessment of speech accuracy was identified. The phonemic scoring method was found to be a more sensitive measure of TE speech accuracy. The W&B list yielded the lowest accuracy scores of the three lists. This finding may indicate its increased sensitivity and potential clinical value. Conclusions: Overall, this study supports the use of open-set, phonemic scoring methods when evaluating TE speaker intelligibility. Future research should aim to assess the specificity of assessment tools on a larger sample of TE speakers who vary in their speech proficiency.

Associations between rapid auditory processing of speech sounds and specific verbal communication skills in autism.

The ability to rapidly process speech sounds is integral not only for processing other's speech, but also for auditory processing of one's own speech, which allows for maintenance of speech accuracy. Deficits in rapid auditory processing have been demonstrated in autistic individuals, particularly those with language impairment. We examined rapid auditory processing for speech sounds in relation to performance on a battery of verbal communication measures to determine which aspects of verbal communication were associated with cortical auditory processing in a sample of individuals with autism. Participants were 57 children and adolescents (40 male and 17 female) ages 5-18 who were diagnosed with an Autism Spectrum Disorder (ASD). Rapid auditory processing of speech sounds was measured via a magnetoencephalographic (MEG) index of the quality of the auditory evoked response to the second of two differing speech sounds ("Ga" / "Da") presented in rapid succession. Verbal communication abilities were assessed on standardized clinical measures of overall expressive and receptive language, vocabulary, articulation, and phonological processing. Associations between cortical measures of left- and right-hemisphere rapid auditory processing and verbal communication measures were examined. Rapid auditory processing of speech sounds was significantly associated with speech articulation bilaterally (r = 0.463, p = 0.001 for left hemisphere and r = 0.328, p = 0.020 for right hemisphere). In addition, rapid auditory processing in the left hemisphere was significantly associated with overall expressive language abilities (r = 0.354, p = 0.013); expressive (r = 0.384, p = 0.005) vocabulary; and phonological memory (r = 0.325, p = 0.024). Phonological memory was found to mediate the relationship between rapid cortical processing and receptive language. These results demonstrate that impaired rapid auditory processing for speech sounds is associated with dysfunction in verbal communication in ASD. The data also indicate that intact rapid auditory processing may be necessary for even basic communication skills that support speech production, such as phonological memory and articulatory control.