Saved in:
Bibliographic Details
Main Authors: Fang, Yu-Hsuan, Lo, Tien-Hong, Sung, Yao-Ting, Chen, Berlin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.12591
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912541377036288
author Fang, Yu-Hsuan
Lo, Tien-Hong
Sung, Yao-Ting
Chen, Berlin
author_facet Fang, Yu-Hsuan
Lo, Tien-Hong
Sung, Yao-Ting
Chen, Berlin
contents Traditional Automated Speaking Assessment (ASA) systems exhibit inherent modality limitations: text-based approaches lack acoustic information while audio-based methods miss semantic context. Multimodal Large Language Models (MLLM) offer unprecedented opportunities for comprehensive ASA by simultaneously processing audio and text within unified frameworks. This paper presents a very first systematic study of MLLM for comprehensive ASA, demonstrating the superior performance of MLLM across the aspects of content and language use . However, assessment on the delivery aspect reveals unique challenges, which is deemed to require specialized training strategies. We thus propose Speech-First Multimodal Training (SFMT), leveraging a curriculum learning principle to establish more robust modeling foundations of speech before cross-modal synergetic fusion. A series of experiments on a benchmark dataset show MLLM-based systems can elevate the holistic assessment performance from a PCC value of 0.783 to 0.846. In particular, SFMT excels in the evaluation of the delivery aspect, achieving an absolute accuracy improvement of 4% over conventional training approaches, which also paves a new avenue for ASA.
format Preprint
id arxiv_https___arxiv_org_abs_2508_12591
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Beyond Modality Limitations: A Unified MLLM Approach to Automated Speaking Assessment with Effective Curriculum Learning
Fang, Yu-Hsuan
Lo, Tien-Hong
Sung, Yao-Ting
Chen, Berlin
Computation and Language
Artificial Intelligence
Sound
Traditional Automated Speaking Assessment (ASA) systems exhibit inherent modality limitations: text-based approaches lack acoustic information while audio-based methods miss semantic context. Multimodal Large Language Models (MLLM) offer unprecedented opportunities for comprehensive ASA by simultaneously processing audio and text within unified frameworks. This paper presents a very first systematic study of MLLM for comprehensive ASA, demonstrating the superior performance of MLLM across the aspects of content and language use . However, assessment on the delivery aspect reveals unique challenges, which is deemed to require specialized training strategies. We thus propose Speech-First Multimodal Training (SFMT), leveraging a curriculum learning principle to establish more robust modeling foundations of speech before cross-modal synergetic fusion. A series of experiments on a benchmark dataset show MLLM-based systems can elevate the holistic assessment performance from a PCC value of 0.783 to 0.846. In particular, SFMT excels in the evaluation of the delivery aspect, achieving an absolute accuracy improvement of 4% over conventional training approaches, which also paves a new avenue for ASA.
title Beyond Modality Limitations: A Unified MLLM Approach to Automated Speaking Assessment with Effective Curriculum Learning
topic Computation and Language
Artificial Intelligence
Sound
url https://arxiv.org/abs/2508.12591