Nina R Benway, Jonathan L Preston, Asif Salekin, Elaine Hitchcock, Tara McAllister
{"title":"评估声学表征和归一化对语言发音障碍儿童的翘舌音分类。","authors":"Nina R Benway, Jonathan L Preston, Asif Salekin, Elaine Hitchcock, Tara McAllister","doi":"10.1121/10.0024632","DOIUrl":null,"url":null,"abstract":"<p><p>The effects of different acoustic representations and normalizations were compared for classifiers predicting perception of children's rhotic versus derhotic /ɹ/. Formant and Mel frequency cepstral coefficient (MFCC) representations for 350 speakers were z-standardized, either relative to values in the same utterance or age-and-sex data for typical /ɹ/. Statistical modeling indicated age-and-sex normalization significantly increased classifier performances. Clinically interpretable formants performed similarly to MFCCs and were endorsed for deep neural network engineering, achieving mean test-participant-specific F1-score = 0.81 after personalization and replication (σx = 0.10, med = 0.83, n = 48). Shapley additive explanations analysis indicated the third formant most influenced fully rhotic predictions.</p>","PeriodicalId":73538,"journal":{"name":"JASA express letters","volume":null,"pages":null},"PeriodicalIF":1.2000,"publicationDate":"2024-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11522988/pdf/","citationCount":"0","resultStr":"{\"title\":\"Evaluating acoustic representations and normalization for rhoticity classification in children with speech sound disorders.\",\"authors\":\"Nina R Benway, Jonathan L Preston, Asif Salekin, Elaine Hitchcock, Tara McAllister\",\"doi\":\"10.1121/10.0024632\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><p>The effects of different acoustic representations and normalizations were compared for classifiers predicting perception of children's rhotic versus derhotic /ɹ/. Formant and Mel frequency cepstral coefficient (MFCC) representations for 350 speakers were z-standardized, either relative to values in the same utterance or age-and-sex data for typical /ɹ/. Statistical modeling indicated age-and-sex normalization significantly increased classifier performances. Clinically interpretable formants performed similarly to MFCCs and were endorsed for deep neural network engineering, achieving mean test-participant-specific F1-score = 0.81 after personalization and replication (σx = 0.10, med = 0.83, n = 48). Shapley additive explanations analysis indicated the third formant most influenced fully rhotic predictions.</p>\",\"PeriodicalId\":73538,\"journal\":{\"name\":\"JASA express letters\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":1.2000,\"publicationDate\":\"2024-02-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11522988/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"JASA express letters\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1121/10.0024632\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"ACOUSTICS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"JASA express letters","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1121/10.0024632","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"ACOUSTICS","Score":null,"Total":0}
引用次数: 0
摘要
我们比较了不同声学表征和归一化对预测儿童发音/ɹ/的分类器的影响。对 350 名说话者的声形和梅尔频率倒谱系数(MFCC)表征进行了 z 标准化,或者相对于同一语料中的值,或者相对于典型 /ɹ/ 的年龄和性别数据。统计建模表明,年龄和性别标准化显著提高了分类器的性能。临床可解释声母的表现与 MFCC 相似,并得到了深度神经网络工程的认可,在个性化和复制后,测试参与者特定的平均 F1 分数 = 0.81(σx = 0.10,中间值 = 0.83,n = 48)。夏普利加法解释分析表明,第三声母对完全斜音预测的影响最大。
Evaluating acoustic representations and normalization for rhoticity classification in children with speech sound disorders.
The effects of different acoustic representations and normalizations were compared for classifiers predicting perception of children's rhotic versus derhotic /ɹ/. Formant and Mel frequency cepstral coefficient (MFCC) representations for 350 speakers were z-standardized, either relative to values in the same utterance or age-and-sex data for typical /ɹ/. Statistical modeling indicated age-and-sex normalization significantly increased classifier performances. Clinically interpretable formants performed similarly to MFCCs and were endorsed for deep neural network engineering, achieving mean test-participant-specific F1-score = 0.81 after personalization and replication (σx = 0.10, med = 0.83, n = 48). Shapley additive explanations analysis indicated the third formant most influenced fully rhotic predictions.