The development of an automated method for analyzing communication rate in augmentative and alternative communication.

The development of an automated method for analyzing communication rate in augmentative and alternative communication.
复制标题

开发一种用于分析增强和替代通信中的通信速率的自动化方法。

DOI:
10.1080/10400435.2006.10131910
复制
发表时间:
2006
期刊:
Assistive technology : the official journal of RESNA
影响因子:
--
通讯作者:
Mathy,Pamela
Mathy,Pamela
中科院分区:
--
文献类型:
--
作者:
Smith,LauraE;Higginbotham,DJeffery;Lesher,GregoryW;Moulton,Bryan;Mathy,Pamela

文献摘要

相似文献

在增强和替代交流(AAC)中,循证实践的一个重大障碍是缺乏经验证的性能指标,可供语音语言病理学家和康复工程师用于评估AAC消费者的交流和设备使用。最近,已经做出努力来开发自动数据记录技术,以促进AAC扬声器的设备使用的转录和分析。通信速率的自动测量的主要误差源是存在过多的选择间间隔(ISI)(即,暂停时间),对于这些时间没有发生通信活动。本研究的目标是开发一种自动化技术来过滤极端ISI,同时保持真实的通信速率性能。AAC数据日志文件是从参与AAC技术的1个月现场试验的7个人处获得的。两个时间滤波技术(任意和个人)进行了比较,以消除过多的ISI的能力。结果表明,使用个性化的时间过滤器是更敏感的性能变化比使用任意的时间过滤器。此外,个性化的时间过滤器提高了参与者的通信速率(以每分钟的单词数来衡量),比未过滤的通信速率估计值大1.8到34.5倍。此外,第一AAC通信速率性能的估计,从现场。进一步的研究和有效使用自动数据记录和分析的影响进行了讨论。
A significant barrier to evidence-based practice in Augmentative and Alternative Communication (AAC) is the lack of validated performance measures that can be used by speech-language pathologists and rehabilitation engineers to evaluate the communication and device use of AAC consumers. Recently an effort has been made to develop automated data-logging techniques to facilitate the transcription and analysis of the AAC speaker's device use. A major source of error for the automated measurement of communication rate is the presence of excessive Inter-Selection Intervals (ISIs) (i.e., pause times), for which no communicative activity is occurring. The goal of this study was to develop an automated technique to filter out extreme ISIs, while preserving true communication rate performance. AAC data log files were obtained from seven individuals participating in a 1-month field trial of an AAC technology. Two temporal filtering techniques (arbitrary and individual) were compared for their ability to eliminate excessive ISIs. Results indicated that use of an individualized temporal filter was more sensitive to performance variability than use of an arbitrary temporal filter. Further, the individualized temporal filter elevated the participants' communication rate (measured by words per minute) by a factor of 1.8 to 34.5 greater than that of the unfiltered communication rate estimate. In addition, the first AAC communication rate performance estimates taken from the field are presented. Implications for further research and the valid use of automated data logging and analysis are discussed.