人工智慧提升輔助與替代溝通系統情緒表達之初步研究：涵蓋自閉症與非自閉症兒童

// Artificial Intelligence for Enhancing Emotion Expression in Augmentative and Alternative Communication Systems: A Preliminary Study Including Children Both With and Without Autism

本研究探討將深度學習臉部情緒識別技術DeepFace整合至輔助溝通系統(AAC)，提升情緒表達能力｜12名神經典型兒童與4名自閉症兒童參與，並進行個別化校準以調整系統｜結果顯示特定情緒識別準確度佳且使用者反應正面，強調個性化調整的重要性，未來將擴大多元族群研究以優化情緒感知AAC技術。

// This study explored integrating facial emotion recognition using DeepFace into augmentative and alternative communication (AAC) systems to improve emotional expression. Twelve neurotypical and four autistic children participated in tasks with individualized calibration to tailor the system. Results showed promising accuracy for certain emotions and positive user engagement, highlighting the importance of personalization. The study suggests further research with diverse users and improved data to advance emotion-aware AAC technology.

Published 2026年6月8日