Automatic assessment of spoken language proficiency of non-native\n children

2019/03/15 by Roberto Gretter, Gretter, Roberto, Katharina Allgaier +5
Computer Science · Psychology · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Phonetics and Phonology Research #Speech Recognition and Synthesis #Speech and dialogue systems

paper · pdf · doi:10.48550/arxiv.1903.06409

openalex publication_date 2019/03/15 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28

Abstract

This paper describes technology developed to automatically grade Italian\nstudents (ages 9-16) on their English and German spoken language proficiency.\nThe students' spoken answers are first transcribed by an automatic speech\nrecognition (ASR) system and then scored using a feedforward neural network\n(NN) that processes features extracted from the automatic transcriptions.\nIn-domain acoustic models, employing deep neural networks (DNNs), are derived\nby adapting the parameters of an original out of domain DNN.\n

Related

• TLT-school: a Corpus of Non Native Children Speech
• Non-native children speech recognition through transfer learning
• Non-native Children's Automatic Speech Assessment Challenge (NOCASA)
• Error-preserving Automatic Speech Recognition of Young English Learners' Language
• The NTNU System at the Interspeech 2020 Non-Native Children's Speech ASR Challenge