Automatic GRBAS Scoring of Pathological Voices using Deep Learning and a Small Set of Labeled Voice Data

Hidaka, Shunsuke; Lee, Yogaku; Nakanishi, Moe; Wakamiya, Kohei; Nakagawa, Takashi; Kaburagi, Tokihiko

doi:10.1016/j.jvoice.2022.10.020

Cited by 6 publications

(1 citation statement)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The grade-roughness-breathiness-asthenia-strain (GRBAS) scale is a perceptual subjective voice assessment to evaluate pathological voice quality [ 16 ]. We used the GRBAS of the Japan Society of Logopedics and Phoniatrics to evaluate the vocal disorders in the 2 groups, including grade, roughness, breathiness, asthenia, and strain.…”

Section: Methodsmentioning

confidence: 99%

Effects of Conventional Speech Therapy with Liuzijue Qigong, a Traditional Chinese Method of Breath Training, in 70 Patients with Post-Stroke Spastic Dysarthria

et al. 2023

View full text Add to dashboard Cite

Background Post-stroke spastic dysarthria (PSSD) is a motor speech impairment that impacts patient communication and quality of life. Liuzijue Qigong (LQG), a traditional Chinese method of breath training, could serve as an effective treatment for PSSD. This study compared the effects of conventional speech therapy and conventional speech therapy combined with LQG in patients with PSSD. Material/Methods Seventy patients with PSSD were randomly divided into a control group (conventional speech therapy, n=35, 77.14% cerebral infarction, 22.86% cerebral hemorrhage) and experimental group (LQG combined with conventional speech therapy, n=35, 85.71% cerebral infarction, 14.29% cerebral hemorrhage). Conventional speech therapy included relaxation, breath control, organ articulation, and pronunciation training. LQG involved producing 6 different sounds (Xu, He, Hu, Si, Chui, and Xi) accompanied by breathing and body movements. Patients were treated once a day, 5 times a week, for 4 weeks. The Frenchay Dysarthria Assessment scale (FDA), speech articulation, maximum phonation time (MPT), loudness, and Montreal Cognitive Assessment scale (MoCA) were evaluated. Results At 4 weeks, the experimental group showed significant improvements compared with the control group in the change of FDA (13.26±6.84 vs 18.03±5.32, P =0.028), speech articulation (63.17±22.40 vs 76.51±15.28, P =0.024), MPT (1.34±1.30 vs 3.89±3.98, P <0.001), loudness (3.46±2.74 vs 7.14±2.56, P =0.009), MoCA (19.40±3.72 vs 22.20±5.30, P =0.020), total effective rate (68.57% vs 88.57%, P =0.041). Conclusions LQG, when combined with conventional speech therapy, enhanced the comprehensive speech ability of patients with PSSD compared with conventional treatment alone.

show abstract

Section: Methodsmentioning

confidence: 99%