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

The present acoustic experiment studies the acquisition of Standard Chinese vowels by native speakers of Uyghur. The results indicate that in identifying the 7 vowels of Standard Chinese produced by the native Han Chinese speakers, Uyghur subjects with high Chinese proficiency perform better both in accuracy and reaction time than Uyghur subjects with low Chinese proficiency, and all the Uyghur subjects are not good in perceiving vowels //, /i/ and //. The results of the experiment support the Speech Learning Model and the Perceptual Assimilation Model. In addition, it shows that the second language speech acquisition is restricted by the first language phonological system.

Download Full-text

Auditory-orthographic integration at the onset of L2 speech acquisition

Language and Speech ◽

10.1177/0023830918777537 ◽

2018 ◽

Vol 62 (3) ◽

pp. 427-451 ◽

Cited By ~ 1

Author(s):

Yasaman Rafat ◽

Ryan A Stevenson

Keyword(s):

Second Language ◽

First Language ◽

Training Condition ◽

Speech Acquisition ◽

Negative Effect ◽

English Speaking ◽

Spanish Phonology ◽

Speech Learning

Recent studies have provided evidence for both a positive and a negative effect of orthography on second language speech learning. However, not much is known about whether orthography can trigger a McGurk-like effect (McGurk & MacDonald, 1976) in second language speech learning. This study examined whether exposure to auditory and orthographic input may lead to a McGurk-like effect in naïve English-speaking participants learning a second language with Spanish phonology and orthography. Specifically, it reports on (a) production of non-target-like combinations such as [lj] as in [poljo] for <pollo>-[pojo], where the auditory Spanish [j] and the first language English [l] that correspond to the shared digraph <ll> are integrated, and (b) fusion quantified in terms of [z] devoicing such as [z̥apito] for <zapito>-[zapito]. Moreover, the effects of (a) type of grapheme-to-sound correspondence, (b) position in the word, and (c) condition of training and testing were examined. Participants were assigned to four groups: (a) auditory only, (b) orthography at training and production, (c) orthography at training, and (d) orthography at production. The positions included word-initial and word-medial. The grapheme-to-sound correspondences consisted of <v>-[b], <d>-[δ], <z>-[s] and <ll>-[j]. Results were indicative of a McGurk-like effect only for the Spanish digraph <ll>. The highest rate of combination productions was attested in the orthography-training condition in the word-medial position.

Download Full-text

Modeling the Development of Pronunciation in Infant Speech Acquisition

Motor Control ◽

10.1123/mcj.15.1.85 ◽

2011 ◽

Vol 15 (1) ◽

pp. 85-117 ◽

Cited By ~ 36

Author(s):

Ian S. Howard ◽

Piers Messum

Keyword(s):

Computational Model ◽

Speech Sound ◽

Equivalence Relations ◽

Speech Sounds ◽

Speech Acquisition ◽

Part Of Speech ◽

Vocal Apparatus ◽

Infant Speech

Pronunciation is an important part of speech acquisition, but little attention has been given to the mechanism or mechanisms by which it develops. Speech sound qualities, for example, have just been assumed to develop by simple imitation. In most accounts this is then assumed to be by acoustic matching, with the infant comparing his output to that of his caregiver. There are theoretical and empirical problems with both of these assumptions, and we present a computational model—Elija—that does not learn to pronounce speech sounds this way. Elija starts by exploring the sound making capabilities of his vocal apparatus. Then he uses the natural responses he gets from a caregiver to learn equivalence relations between his vocal actions and his caregiver’s speech. We show that Elija progresses from a babbling stage to learning the names of objects. This demonstrates the viability of a non-imitative mechanism in learning to pronounce.

Download Full-text

Background II: Infant Speech Acquisition

Phonological Acquisition ◽

10.1007/978-1-137-54306-6_2 ◽

2016 ◽

pp. 29-50

Author(s):

Anne-Michelle Tessier

Keyword(s):

Speech Acquisition ◽

Infant Speech

Download Full-text

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

10.31234/osf.io/t7anp ◽

2021 ◽

Author(s):

Jessie S. Nixon ◽

Fabian Tomaschek

Keyword(s):

Speech Perception ◽

Speech Signal ◽

A Priori ◽

Speech Sound ◽

Model Performance ◽

High Amplitude ◽

Clustering Methods ◽

Early Infant ◽

Infant Speech ◽

Statistical Clustering

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 mod- elling early infant speech sound acquisition.

Download Full-text

Parents' Perception of Diagnostic Reports and Conferences

Language Speech and Hearing Services in Schools ◽

10.1044/0161-1461.1402.114 ◽

1983 ◽

Vol 14 (2) ◽

pp. 114-120 ◽

Cited By ~ 2

Author(s):

Betty U. Watson ◽

Ronald W. Thompson

Keyword(s):

Medical Treatment ◽

Diagnostic Information ◽

Learning Language ◽

Speech Learning ◽

Hearing Problems

The purpose of this study was to evaluate parents' reactions and understanding of diagnostic information from written reports and conferences in a clinic which provides multidisciplinary evaluations for children with speech, learning, language, and hearing problems. Previous studies and anecdotal reports suggested that many parents do not receive appropriate diagnostic information about their children. In the present study questionnaires were mailed to parents who had received reports of evaluations and most of whom had attended hour-long conferences covering the findings. Questionnaires were also sent to professionals who had received reports. Fifty-seven percent of the parents, and 63% of the professionals returned the questionnaires. Ninety percent of the parents indicated that they had understood the results as they were presented in the conference. Ninety-three percent of the professionals and 89% of the parents stated they understood the conclusions of the written reports .Further, 83% of the parents and 80% of the professionals reported that the findings had made a change in the child's educational or medical treatment. The percentage of parents who reported understanding the findings was greater than expected. The specific informing techniques used in this study are discussed.

Download Full-text

Classification of Infant Vocalizations by Untrained Listeners

Journal of Speech Language and Hearing Research ◽

10.1044/2019_jslhr-s-18-0494 ◽

2019 ◽

Vol 62 (9) ◽

pp. 3265-3275

Author(s):

Heather L. Ramsdell-Hudock ◽

Anne S. Warlaumont ◽

Lindsey E. Foss ◽

Candice Perry

Keyword(s):

Formal Education ◽

Additional Insight ◽

Listener Responses ◽

Infant Vocalization ◽

Early Infant ◽

Infant Vocalizations ◽

Audio Recordings ◽

Speech And Language Development ◽

Insight Into

Purpose To better enable communication among researchers, clinicians, and caregivers, we aimed to assess how untrained listeners classify early infant vocalization types in comparison to terms currently used by researchers and clinicians. Method Listeners were caregivers with no prior formal education in speech and language development. A 1st group of listeners reported on clinician/researcher-classified vowel, squeal, growl, raspberry, whisper, laugh, and cry vocalizations obtained from archived video/audio recordings of 10 infants from 4 through 12 months of age. A list of commonly used terms was generated based on listener responses and the standard research terminology. A 2nd group of listeners was presented with the same vocalizations and asked to select terms from the list that they thought best described the sounds. Results Classifications of the vocalizations by listeners largely overlapped with published categorical descriptors and yielded additional insight into alternate terms commonly used. The biggest discrepancies were found for the vowel category. Conclusion Prior research has shown that caregivers are accurate in identifying canonical babbling, a major prelinguistic vocalization milestone occurring at about 6–7 months of age. This indicates that caregivers are also well attuned to even earlier emerging vocalization types. This supports the value of continuing basic and clinical research on the vocal types infants produce in the 1st months of life and on their potential diagnostic utility, and may also help improve communication between speech-language pathologists and families.

Download Full-text

Editorial Note

Journal of Speech Disorders ◽

10.1044/jshd.1101.02 ◽

1946 ◽

Vol 11 (1) ◽

pp. 2-2

Keyword(s):

Editorial Note ◽

Speech Sounds ◽

Left Hand ◽

Hand Column ◽

Right Hand ◽

Infant Speech ◽

The Right

In the article “Infant Speech Sounds and Intelligence” by Orvis C. Irwin and Han Piao Chen, in the December 1945 issue of the Journal, the paragraph which begins at the bottom of the left hand column on page 295 should have been placed immediately below the first paragraph at the top of the right hand column on page 296. To the authors we express our sincere apologies.

Download Full-text