Multilingual Speech Evaluation: Case Studies on English, Malay and Tamil
This work addresses the problem of computer-assisted language learning for low-resource languages, offering a more generalizable approach compared to monolingual designs, though it is incremental in applying existing techniques to new domains.
The paper tackled the challenge of automatic speech evaluation for low-resource languages by developing a method using music-inspired features and vector representations, achieving consistent gains in predicting pronunciation, rhythm, and intonation for English, Malay, and Tamil.
Speech evaluation is an essential component in computer-assisted language learning (CALL). While speech evaluation on English has been popular, automatic speech scoring on low resource languages remains challenging. Work in this area has focused on monolingual specific designs and handcrafted features stemming from resource-rich languages like English. Such approaches are often difficult to generalize to other languages, especially if we also want to consider suprasegmental qualities such as rhythm. In this work, we examine three different languages that possess distinct rhythm patterns: English (stress-timed), Malay (syllable-timed), and Tamil (mora-timed). We exploit robust feature representations inspired by music processing and vector representation learning. Empirical validations show consistent gains for all three languages when predicting pronunciation, rhythm and intonation performance.