Big data suggest strong constraints of linguistic similarity on adult language learning

Schepens, Job; Hout, R.W.N.M. van; Jaeger, T. Florian

doi:10.1016/j.cognition.2019.104056

Cited by 39 publications

(50 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The development prospects of artificial intelligence language training institutions are excellent [35][36][37]. Online language education directly hits the pain points of the industry.…”

Section: Scenariomentioning

confidence: 99%

Generative Adversarial Network-Based Neural Audio Caption Model for Oral Evaluation

et al. 2020

View full text Add to dashboard Cite

Oral evaluation is one of the most critical processes in children’s language learning. Traditionally, the Scoring Rubric is widely used in oral evaluation for providing a ranking score by assessing word accuracy, phoneme accuracy, fluency, and accent position of a tester. In recent years, by the emerging demands of the market, oral evaluation requires not only providing a single score from pronunciation but also in-depth, meaning comments based on content, context, logic, and understanding. However, the Scoring Rubric requires massive human work (oral evaluation experts) to provide such deep meaning comments. It is considered uneconomical and inefficient in the current market. Therefore, this paper proposes an automated expert comment generation approach for oral evaluation. The approach first extracts the oral features from the children’s audio as well as the text features from the corresponding expert comments. Then, a Gated Recurrent Unit (GRU) is applied to encode the oral features into the model. Afterwards, a Long Short-Term Memory (LSTM) model is applied to train the mappings between oral features and text features and generate expert comments for the new coming oral audio. Finally, a Generative Adversarial Network (GAN) is combined to improve the quality of the generated comments. It generates pseudo-comments to train the discriminator to recognize the human-like comments. The proposed approach is evaluated in a real-world audio dataset (children oral audio) collected by our collaborative company. The proposed approach is also integrated into a commercial application to generate expert comments for children’s oral evaluation. The experimental results and the lessons learned from real-world applications show that the proposed approach is effective for providing meaningful comments for oral evaluation.

show abstract

“…The development prospects of artificial intelligence language training institutions are excellent [35][36][37]. Online language education directly hits the pain points of the industry.…”

Section: Scenariomentioning

confidence: 99%

Generative Adversarial Network-Based Neural Audio Caption Model for Oral Evaluation

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The (dis)similarity of L1 and L2 sounds determines the perception of L2 categories in inexperienced listeners (see the Perceptual Assimilation model as described in Best, 1995;Best and Tyler, 2007;Grimaldi et al, 2014) as well as experienced listeners (see the Speech Learning model as described in Flege, 1995;Flege, 2003). Beyond phonological category learning, a recent large-scale study by Schepens et al (2020) examined the role of L1 on L2 learning in 50,000 Dutch learners of 62 diverse L1s using language-testing scores that assess speaking, writing, reading, and listening proficiency. The study found that L1 exerts a huge impact on L2 learning through the phonological, morphological and lexical similarities between the L1s and the L2.…”

Section: All As Second Language Acquisitionmentioning

confidence: 99%

Modelling L1 and the artificial language during artificial language learning

Tang¹,

Baer-Henney²

2020

Preprint

View full text Add to dashboard Cite

Artificial language learning research (ALL) revealed that learners show bias towards certain artificial language patterns which they acquired with relative ease as a result of a universal mechanism that drives the learning process of a natural language. Building on insights from second language acquisition and psycholinguistic research in lexical processing, this study filled an important knowledge gap in ALL research by demonstrating the need to model the influence of both the L1 and the target language (L2) on learning bias. Specifically, native speakers of German (n = 232) and Mandarin (n = 219) were taught a phonological pattern using a set of non-words. In addition to replicating the impact of a consonant identity bias previously found in English (Linzen and Gallagher, 2017) for two very distinct languages, by training an analogical and a discriminative learning model over the lexicons of the L1 and the L2, we were able to capture their influences in the learning of the L2. Nonwords are more likely to be accepted as grammatical, if they are more similar to the L2 and more different from the L1. In addition, nonwords are less likely to be accepted as grammatical if the decision takes longer and if the nonword is judged at a later time. Our findings show that we are at risk to underestimate the role of the lexicons when examining learning biases with ALL. Some recommendations for conducting cross-linguistic ALL experiments were made.

show abstract

“…It is well testified that a learner's L1 influences the acquisition and use of a new language (e.g., Schepens et al, 2020). As a widely known example from outside academia, the Foreign Service Institute of the United States Department of State has for decades categorized languages in groups based on how long it is estimated that an English L1 speaker needs in order to acquire the L2 to a certain proficiency level (https://www.state.…”

Section: Crosslinguistic Influencementioning

confidence: 99%

Language Aptitude and Crosslinguistic Influence in Initial L2 Learning

Bokander

2020

JESLA

View full text Add to dashboard Cite

Big data suggest strong constraints of linguistic similarity on adult language learning

Cited by 39 publications

References 73 publications

Generative Adversarial Network-Based Neural Audio Caption Model for Oral Evaluation

Generative Adversarial Network-Based Neural Audio Caption Model for Oral Evaluation

Modelling L1 and the artificial language during artificial language learning

Language Aptitude and Crosslinguistic Influence in Initial L2 Learning

Contact Info

Product

Resources

About