scholarly journals Vocabulary Size Influences Spontaneous Speech in Native Language Users: Validating the Use of Automatic Speech Recognition in Individual Differences Research

2020 ◽  
pp. 002383092091107 ◽  
Author(s):  
Suzanne R. Jongman ◽  
Yung Han Khoe ◽  
Florian Hintz
Keyword(s):  
Speech Recognition ◽  
Automatic Speech Recognition ◽  
Language Production ◽  
Spontaneous Speech ◽  
Receptive Vocabulary ◽  
Sentence Production ◽  
Production Data ◽  
Vocabulary Size ◽  
Speech Transcription ◽  
The Relationship

Previous research has shown that vocabulary size affects performance on laboratory word production tasks. Individuals who know many words show faster lexical access and retrieve more words belonging to pre-specified categories than individuals who know fewer words. The present study examined the relationship between receptive vocabulary size and speaking skills as assessed in a natural sentence production task. We asked whether measures derived from spontaneous responses to everyday questions correlate with the size of participants’ vocabulary. Moreover, we assessed the suitability of automatic speech recognition (ASR) for the analysis of participants’ responses in complex language production data. We found that vocabulary size predicted indices of spontaneous speech: individuals with a larger vocabulary produced more words and had a higher speech-silence ratio compared to individuals with a smaller vocabulary. Importantly, these relationships were reliably identified using manual and automated transcription methods. Taken together, our results suggest that spontaneous speech elicitation is a useful method to investigate natural language production and that automatic speech recognition can alleviate the burden of labor-intensive speech transcription.

Are Communicative Gestures the Substrate of Language?

1997 ◽  
Vol 1 (2) ◽  
pp. 261-282 ◽  
Author(s):  
Joanna Blake ◽  
Esther Olshansky ◽  
Grace Vitale ◽  
Silvana Macdonald
Keyword(s):  
Language Production ◽  
Communicative Competence ◽  
Free Play ◽  
Receptive Vocabulary ◽  
Vocabulary Size ◽  
Second Year ◽  
Language Measures ◽  
Communicative Gestures ◽  
Relationship Of ◽  
The Relationship

The relationship between communicative gestures and language acquisition was investigated in 30 infants who were visited at home four times between 9 months and 3 years. At 9 and 15 months, they were videotaped in free play with their mothers, and their communicative gestures were coded from these interactions. At three years, measures of spontaneous speech, receptive vocabulary, and communicative competence were obtained. So-called primitive gestures, Protest/Rejection and some forms of Request, were found to decline in the second year and to be negatively related to measures of language production at 3 years. Object Exchange and Comment gestures, in contrast, increased in the second year and were positively related to language measures, the first to early vocabulary size and the second to 3-year-old receptive vocabulary. Gesturing as a whole did not decline with the onset of language, and the co-ordination of gestures with vocalizations increased. These findings support both a precursor model and an interdependent model of the relationship of gestures to language.

Speech and Language Production at Age 2

1991 ◽  
Vol 34 (4) ◽  
pp. 879-892 ◽  
Author(s):  
Lauren K. Nelson ◽  
Harold R. Bauer
Keyword(s):  
Language Acquisition ◽  
Language Processing ◽  
Language Production ◽  
Complexity Analysis ◽  
Spontaneous Speech ◽  
Limited Capacity ◽  
Speech And Language ◽  
Lexical Items ◽  
Propositional Complexity ◽  
The Relationship

The purpose of this study was to explore how 2-year-old children manage the relationship between phonetic production and production of word combinations in their spontaneous speech. The subjects were 5 normally developing 2-year-olds who were participants in an ongoing longitudinal study of speech and language acquisition. Three measures were used to estimate phonetic production skills in the children’s spontaneous speech samples. These included a measure of the accuracy of consonant production (Percentage of Consonants Correct), and two estimators of phonetic complexity (phonetic products for utterance and word length units). Regression analyses were used to determine the relationship between complexity of word combinations, as measured by length of utterance in morphemes and a propositional complexity analysis, and utilization of phonetic production skills. The results revealed modest tradeoffs between complexity of word combinations and accuracy of consonant production for 2 of the 5 children. The results also showed tradeoffs between complexity of word combinations and phonetic complexity of individual lexical items (phonetic product for words) for 4 of the 5 children. As the complexity of these 4 children’s multiword combinations increased, the phonetic complexity of individual lexical items decreased. These results are consistent with synergistic theories of language acquisition and language processing that emphasize dynamic tradeoffs in interactions among language processing levels in a limited capacity production system.

scholarly journals Does the Use of Complex Sentences Differentiate Between Bilinguals With and Without DLD? Evidence From Conversation and Narrative Tasks

2022 ◽  
Vol 6 ◽  
Author(s):  
Johanne Paradis ◽  
Tamara Sorenson Duncan ◽  
Stephanie Thomlinson ◽  
Brian Rusk
Keyword(s):  
Relative Clauses ◽  
Language Disorder ◽  
Receptive Vocabulary ◽  
Sentence Production ◽  
Bilingual Children ◽  
School Age ◽  
Complex Sentence ◽  
Vocabulary Size ◽  
Complex Sentences ◽  
Adverbial Clauses

Over-identification of language disorder among bilingual children with typical development (TD) is a risk factor in assessment. One strategy for improving assessment accuracy with bilingual children is to determine which linguistic sub-domains differentiate bilingual children with TD from bilingual children with developmental language disorder (DLD). To date, little research on sequential bilinguals with TD and DLD has focussed on complex (multi-clausal) sentences in naturalistic production, even though this is a noted domain of weakness for school-age monolinguals with DLD. Accordingly, we sought to determine if there were differences in the use of complex sentences in conversational and narrative tasks between school-age sequential bilinguals with TD and with DLD at the early stages of L2 acquisition. We administered a conversation and a narrative task to 63 English L2 children with TD and DLD, aged 5–7 years with 2 years of exposure to the L2. Children had diverse first language backgrounds. The L2-TD and L2-DLD groups were matched for age, length of L2 exposure and general L2 proficiency (receptive vocabulary size). Language samples from both tasks were coded and analyzed for the use of complex versus simple sentences, for the distribution of complex sentence types, for clausal density and mean length of utterance (MLU). Complex sentences included coordinated clauses, sentential complement clauses, adverbial clauses and relative clauses. Using regression modelling and PERMANOVA, we found that the L2-TD group produced more complex sentences than the L2-DLD group, with coordinated clauses, adverbial clauses and relative clauses differing the most between the groups. Furthermore, the two groups differed for mean clausal density, but not for MLU, indicating that clausal density and MLU did not estimate identical morphosyntactic abilities. Individual variation in complex sentence production for L2-TD was predicted by longer L2 exposure and task; by contrast, for L2-DLD, it was predicted by older age. This study indicates that complex sentence production is an area of weakness for bilingual children with DLD, as it is for monolinguals with DLD. The clinical implications of these findings are discussed.

To ‘errrr’ is human: ecology and acoustics of speech disfluencies

2001 ◽  
Vol 31 (1) ◽  
pp. 153-169 ◽  
Author(s):  
Elizabeth Shriberg
Keyword(s):  
Automatic Speech Recognition ◽  
Human Ecology ◽  
Language Production ◽  
American English ◽  
Voice Quality ◽  
Spontaneous Speech ◽  
Human Language ◽  
Vowel Quality ◽  
Task Effects ◽  
Filled Pauses

Unlike read or laboratory speech, spontaneous speech contains high rates of disfluencies (e.g. repetitions, repairs, filled pauses, false starts). This paper aims to promote ‘disfluency awareness’ especially in the field of phonetics –which has much to offer in the way of increasing our understanding of these phenomena. Two broad claims are made, based on analyses of disfluencies in different corpora of spontaneous American English speech. First, an Ecology Claim suggests that disfluencies are related to aspects of the speaking environments in which they arise. The claim is supported by evidence from task effects, location analyses, speaker effects and sociolinguistic effects. Second, an Acoustics Claim argues that disfluency has consequences for phonetic and prosodic aspects of speech that are not represented in the speech patterns of laboratory speech. Such effects include modifications in segment durations, intonation, voice quality, vowel quality and coarticulation patterns. The ecological and acoustic evidence provide insights about human language production in real-world contexts. Such evidence can also guide methods for the processing of spontaneous speech in automatic speech recognition applications.

Case and agreement in English language development

1999 ◽  
Vol 26 (2) ◽  
pp. 357-372 ◽  
Author(s):  
MATTHEW RISPOLI
Keyword(s):  
English Language ◽  
Primary Caregivers ◽  
English Language Development ◽  
Sentence Production ◽  
Production Data ◽  
Sentence Level ◽  
Retrieval Factor ◽  
Third Person ◽  
Subject Pronoun ◽  
The Relationship

This study examines the relationship between third person singular (3Psg) subject pronoun case and agreement, focusing on the hypothesis that these two grammatical subsystems develop together. This hypothesis is broken down into two separate, empirically testable hypotheses: (a) that correct subject case pronoun production and the production of agreement are correlated, and (b) that at the sentence level, correct case is dependent on the presence of agreement. Twenty-nine children between the ages of 2;6 and 4;0 were each audiotaped for approximately two hours playing and interacting with their primary caregivers. Transcribed production data showed that 3Psg masculine subject pronoun case was correlated with agreement marking, whereas 3Psg feminine subject pronoun case was not. This result suggests the influence of a retrieval factor, termed the DOUBLE-CELL EFFECT, on the her for she pronoun case error. At the utterance level, pronoun case was independent of the presence of agreement. Overall, the study indicates that the relationship between case and agreement may be discernible as a general correlation, yet indiscernible at the level of sentence production.

Evaluation of an Automatic Speech Recognition Platform for Dysarthric Speech

2020 ◽  
pp. 1-10
Author(s):  
Irene Calvo ◽  
Peppino Tropea ◽  
Mauro Viganò ◽  
Maria Scialla ◽  
Agnieszka B. Cavalcante ◽  
...  
Keyword(s):  
Speech Recognition ◽  
Technology Acceptance ◽  
Automatic Speech Recognition ◽  
Recognition Accuracy ◽  
Ease Of Use ◽  
Clinical Environment ◽  
Dysarthric Speech ◽  
Acceptance Model ◽  
The Relationship ◽  
Accuracy Rates

<b><i>Introduction:</i></b> The use of commercially available automatic speech recognition (ASR) software is challenged when dysarthria accompanies a physical disability. To overcome this issue, a mobile and personal speech assistant (mPASS) platform was developed, using a speaker-dependent ASR software. <b><i>Objective:</i></b> The aim of this study was to evaluate the performance of the proposed platform and to compare mPASS recognition accuracy to a commercial speaker-independent ASR software. In addition, secondary aims were to investigate the relationship between severity of dysarthria and accuracy and to explore people with dysarthria perceptions on the proposed platform. <b><i>Methods:</i></b> Fifteen individuals with dysarthric speech and 20 individuals with nondysarthric speech recorded 24 words and 5 sentences in a clinical environment. Differences in recognition accuracy between the two systems were evaluated. In addition, mPASS usability was assessed with a technology acceptance model (TAM) questionnaire. <b><i>Results:</i></b> In both groups, mean accuracy rates were significantly higher with mPASS compared to the commercial ASR for words and for sentences. mPASS reached good levels of usefulness and ease of use according to the TAM questionnaire. <b><i>Conclusions:</i></b> Practical applicability of this technology is realistic: the mPASS platform is accurate, and it could be easily used by individuals with dysarthria.

scholarly journals KsponSpeech: Korean Spontaneous Speech Corpus for Automatic Speech Recognition

2020 ◽  
Vol 10 (19) ◽  
pp. 6936 ◽  
Author(s):  
Jeong-Uk Bang ◽  
Seung Yun ◽  
Seung-Hi Kim ◽  
Mu-Yeol Choi ◽  
Min-Kyu Lee ◽  
...  
Keyword(s):  
Speech Recognition ◽  
Automatic Speech Recognition ◽  
Large Scale ◽  
Open Data ◽  
Spontaneous Speech ◽  
Open Domain ◽  
Speech Corpus ◽  
Clean Environment ◽  
End To End ◽  
Repeated Words

This paper introduces a large-scale spontaneous speech corpus of Korean, named KsponSpeech. This corpus contains 969 h of general open-domain dialog utterances, spoken by about 2000 native Korean speakers in a clean environment. All data were constructed by recording the dialogue of two people freely conversing on a variety of topics and manually transcribing the utterances. The transcription provides a dual transcription consisting of orthography and pronunciation, and disfluency tags for spontaneity of speech, such as filler words, repeated words, and word fragments. This paper also presents the baseline performance of an end-to-end speech recognition model trained with KsponSpeech. In addition, we investigated the performance of standard end-to-end architectures and the number of sub-word units suitable for Korean. We investigated issues that should be considered in spontaneous speech recognition in Korean. KsponSpeech is publicly available on an open data hub site of the Korea government.

Sign in / Sign up

Export Citation Format

Share Document