噪声下汉语语句识别表在视, 听, 视+ 听三种模式下的等价性研究.pdfVIP

Chinese Scientific Journal of Hearing and Speech Rehabilitation 临床研究 doi:10.3969/j.issn.1672-4933.2010.04.002 噪声下汉语语句识别表在视、听、视+听 三种模式下的等价性研究* Mandarin sentence recognition in babble test in audio-, visual- and audiovisual modes: List equivalency study 1 2 2 1 1 2 1** ■郗 昕 吴事儒 赵乌兰 冀 飞 陈艾婷 王 枫 韩东一 XI Xin, WU Shi-ru, ZHAO Wu-lan, JI Fei, CHEN Ai-ting , WANG Feng , HAN Dong -yi 验证新编中文普通话版嘈杂语噪声下语句识别表在视、听、视+听三种模式下的等价性。 【摘要】 目的 方法 招募27名听力正常人，采用二因素（视听模式、测试表号）重复测量实验设计，分别在单视觉、单听觉 （65 dB SPL言语强度、-6 dB信噪比）及视听结合三种模式下，在声场中完成对27张嘈杂语噪声下语句识别 表的言语识别率测试。三种视听模式及测试表号在所有受试者中均衡排布。 综合三种视听模式的识 结果 别率，27张表的识别率无显著性差异（ =0.929， =0.568）。视、听、视+听三种模式下的语句识别率分别为 F P 33.4±14.1%、40.9±14.4%、91.4±7.3%，两两之间均表现出显著性差异。 结论 视觉信息的补充，可有 效提高噪声下的语句感知能力。新编中文普通话版嘈杂语噪声下语句识别表在视、听、视+听三种模式下均 等价，可用于人工耳蜗植入疗效的评估。 【 】言语识别；视听觉；噪声 关键词 Obj ective 【 】 Abstract To evaluate the equivalence of Mandarin Sentence in Babble test materials in audio, visual and audio visual modes. Methods Twenty-seven adults with normal hearing were recruited to receive the speech recognition score assessment with 27 lists of Mandarin Sentence in Babble test materials in sound field. The experiments were administrated in two-way repeated-measurement design, the subjects were asked to repeat what they heard in visual-only, audio-only (65 dB SPL, -6 dB SNR) and audio visual mode. The audiovisual modes and lists were balanced among all subjects. Results The 27 lists showed no significant difference (F=0.929,P =0.568) among the three modes. The recognition scores in visual, audio and audio visual modes were 33.4±14.1%, 40.9±14.4%, 91.4±7.3% respectively. Post hoc Turkey test showed significant difference