Table 2 - uploaded by Mengjie Qian
Content may be subject to copyright.
Context in source publication
Similar publications
Citations
... All proposed solutions for CALL v1 and v2 relied heavily on the reference grammar file [9,13,14,15,16,17,18]. For example, one of the last year's submissions [15] processed the ASR output and up to 10 entries from the reference grammar file using the doc2vec model. ...
... All proposed solutions for CALL v1 and v2 relied heavily on the reference grammar file [9,13,14,15,16,17,18]. For example, one of the last year's submissions [15] processed the ASR output and up to 10 entries from the reference grammar file using the doc2vec model. Afterwards, they used the word mover distance to get 10 distances representing the student's utterance. ...
... Results for the second edition of the task were presented at an Interspeech 2018 special session, with 18 entries and six papers (https://regulus.unige.ch/ spokencallsharedtask_2ndedition; [9,10,11,12,13,14]). ...
... All proposed solutions for CALL v1 and v2 relied heavily on the reference grammar file [9,13,14,15,16,17,18]. For example, one of the last year's submissions [15] processed the ASR output and up to 10 entries from the reference grammar file using the doc2vec model. ...
... All proposed solutions for CALL v1 and v2 relied heavily on the reference grammar file [9,13,14,15,16,17,18]. For example, one of the last year's submissions [15] processed the ASR output and up to 10 entries from the reference grammar file using the doc2vec model. Afterwards, they used the word mover distance to get 10 distances representing the student's utterance. ...
This paper presents a scoring system that has shown the top result on the text subset of CALL v3 shared task. The presented system is based on text embeddings, namely NNLM~\cite{nnlm} and BERT~\cite{Bert}. The distinguishing feature of the given approach is that it does not rely on the reference grammar file for scoring. The model is compared against approaches that use the grammar file and proves the possibility to achieve similar and even higher results without a predefined set of correct answers. The paper describes the model itself and the data preparation process that played a crucial role in the model training.
In this paper, we are presenting a language learning system which automatically evaluates English speech linguistically and grammatically. The system works by prompting the learner a question in his native language (text+figure) and waiting for his/her spoken response in English. Different types of features were extracted from the response to assess it in terms of language grammar and meaning errors. The universal sentence encoder was used to encode each sentence into 512-dimensional vector to represent the semantic of the response. Also, we propose a binary embedding approach to produce 438 binary features vectors from the student response. To assess the grammatical errors, different features were extracted using a grammar checker tool and part of speech analysis of the response. Finally, the best two DNN-based models have been fused together to enhance the system performance. The best result on the 2018 shared task test dataset is a D-score of 17.11.
