一種新穎的電腦化語音聽力測驗材料建構方法：感知平衡希伯來語詞彙辨識測試的開發

// A Novel Computerized Approach to Constructing Speech Audiometry Materials: Development of a Perceptually Balanced Hebrew Word Recognition Test

本研究開發並驗證一套電腦化方法，製作等價的希伯來語語音辨識測試詞表｜利用Python演算法平衡詞彙難度、語音單位分布及熟悉度｜經正常聽力者測試結果顯示各詞表表現無顯著差異，符合國際標準｜此方法具客觀且可重複性，有助多語言語音測試標準化。

// This study developed and validated a new computerized method to create equivalent Hebrew word lists for speech recognition tests. A custom Python algorithm balanced word difficulty, phonemic distribution, and familiarity to form balanced lists, which were then empirically tested with normal-hearing listeners. The results showed no significant performance differences across lists, confirming their equivalency and compliance with international standards. This method offers an objective, replicable approach for standardized speech recognition test construction in various languages.

Published 2026年6月8日