Speech compression in dysarthria (Utianski et al., 2019)

Purpose: Telemedicine, used to offset disparities in access to speech-language therapy, relies on technology that utilizes compression algorithms to transmit signals efficiently. These algorithms have been thoroughly evaluated on healthy speech; however, the effects of compression algorithms on the intelligibility of disordered speech have not been adequately explored.Method: This case study assessed acoustic and perceptual effects of resampling and speech compression (i.e., transcoding) on the speech of 2 individuals with dysarthria. Forced-choice vowel identification and transcription tasks were utilized, completed by 20 naive undergraduate listeners.Results: Results showed relative improvements and decrements in intelligibility, on various measures, based on the speakers’ acoustic profiles. The transcoding of the speech compression algorithm resulted in an enlarged vowel space area and associated improvements in vowel identification for 1 speaker and a smaller vowel space area and decreased vowel identification for the other speaker. Interestingly, there was an overall decrease in intelligibility in the transcription task in this condition for both speakers.Conclusions: There is a complex interplay between dysarthria and compression algorithms that warrants further exploration. The findings suggest that it is critical to be mindful of apparent changes in intelligibility secondary to compression algorithms necessary for practicing telemedicine.Supplemental Material S1. Confusion matrices for each vowel for each condition for Speaker One, where numbers are raw values and indicate the number of tokens identified out of the corpus of 80 classifications per vowel; target vowels are indicated with phonetic (International Phonetic Alphabet [IPA]) notation; corresponding perceived vowels are indicated with orthographic notation. Supplemental Material S2. Confusion matrices for each vowel for each condition for Speaker Two, where numbers are raw values and indicate the number of tokens identified out of the corpus of 80 classifications per vowel; target vowels are indicated with phonetic (International Phonetic Alphabet [IPA]) notation; corresponding perceived vowels are indicated with orthographic notation.Utianski, R. L., Sandoval, S., Berisha, V., Lansford, K. L., & Liss, J. M. (2019). The effects of speech compression algorithms on the intelligibility of two individuals with dysarthric speech. American Journal of Speech-Language Pathology, 28, 195–203. https://doi.org/10.1044/2018_AJSLP-18-0081

研究背景与目的：远程医疗（Telemedicine）旨在弥补言语语言治疗获取途径的公平性差距，其依赖于利用压缩算法高效传输信号的技术。现有研究已针对健康言语对这类压缩算法开展了全面评估，但压缩算法对障碍性言语可懂度的影响尚未得到充分探索。
研究方法：本案例研究评估了重采样与语音压缩（即转码）对2名构音障碍患者言语的声学与感知学影响。研究采用迫选式元音识别与转录任务，由20名未经过训练的本科生听众完成。
研究结果：结果显示，基于两位说话者的声学特征，不同评估指标下的言语可懂度出现了相对提升与下降。语音压缩转码算法使得1名患者的元音空间面积扩大，其元音识别表现随之提升；而另1名患者的元音空间面积缩小，元音识别表现则出现下降。值得注意的是，在该处理条件下，两位患者的转录任务整体可懂度均有所降低。
研究结论：构音障碍与语音压缩算法之间存在复杂的相互作用，这一问题有待进一步探索。本研究结果表明，在开展远程医疗实践时，必须关注由必要的语音压缩算法所导致的言语可懂度表观变化。
补充材料S1：包含说话者1各条件下每个元音的混淆矩阵，其中数值为原始计数，代表每个元音在80个分类样本中被识别的标记数；目标元音采用语音学（国际音标[IPA]）标注，对应的感知元音采用正字法标注。
补充材料S2：包含说话者2各条件下每个元音的混淆矩阵，其中数值为原始计数，代表每个元音在80个分类样本中被识别的标记数；目标元音采用语音学（国际音标[IPA]）标注，对应的感知元音采用正字法标注。
参考文献：Utianski, R. L.、Sandoval, S.、Berisha, V.、Lansford, K. L.、Liss, J. M.（2019）。语音压缩算法对2名构音障碍患者言语可懂度的影响。《美国言语语言病理学杂志》，28，195–203。https://doi.org/10.1044/2018_AJSLP-18-0081