Improving Hypernasality Estimation with Automatic Speech Recognition in Cleft Palate Speech

被引:0
|
作者
Song, Kaitao [1 ]
Wan, Teng [2 ]
Wang, Bixia [2 ]
Jiang, Huiqiang [1 ]
Qiu, Luna [1 ]
Xu, Jiahang [1 ]
Jiang, Liping [2 ]
Lou, Qun [2 ]
Yang, Yuqing [1 ]
Li, Dongsheng [1 ]
Wang, Xudong [2 ]
Qiu, Lili [1 ]
机构
[1] Microsoft Res, Redmond, WA 98052 USA
[2] Shanghai Jiao Tong Univ, Dept Oral & Craniomaxillofacial Surg, Shanghai Ninth Peoples Hosp, Sch Med, Shanghai, Peoples R China
来源
INTERSPEECH 2022 | 2022年
关键词
Cleft Palate; Hypernasality; Automatic Speech Recognition; ACOUSTIC ANALYSIS; LIP; CHILDREN;
D O I
10.21437/Interspeech.2022-438
中图分类号
O42 [声学];
学科分类号
070206 ; 082403 ;
摘要
Hypernasality is an abnormal resonance in human speech production, especially in patients with craniofacial anomalies such as cleft palate. In clinical application, hypernasality estimation is crucial in cleft palate diagnosis, as its results determine the subsequent surgery and additional speech therapy. Therefore, designing an automatic hypernasality assessment method will facilitate speech-language pathologists to make precise diagnoses. Existing methods for hypernasality estimation only conduct acoustic analysis based on low-resource cleft palate dataset, by using statistical or neural network-based features. In this paper, we propose a novel approach that uses automatic speech recognition model to improve hypernasality estimation. Specifically, we first pre-train an encoder-decoder framework in an automatic speech recognition (ASR) objective by using speech-to-text dataset, and then fine-tune ASR encoder on the cleft palate dataset for hypernasality estimation. Benefiting from such design, our model for hypernasality estimation can enjoy the advantages of ASR model: 1) compared with low-resource cleft palate dataset, the ASR task usually includes large-scale speech data in the general domain, which enables better model generalization; 2) the text annotations in ASR dataset guide model to extract better acoustic features. Experimental results on two cleft palate datasets demonstrate that our method achieves superior performance compared with previous approaches.
引用
收藏
页码:4820 / 4824
页数:5
相关论文
共 50 条
  • [1] Automatic Evaluation of Hypernasality and Speech Intelligibility for Children with Cleft Palate
    He, Ling
    Zhang, Jing
    Liu, Qi
    Yin, Heng
    Lech, Margaret
    PROCEEDINGS OF THE 2013 IEEE 8TH CONFERENCE ON INDUSTRIAL ELECTRONICS AND APPLICATIONS (ICIEA), 2013, : 220 - 223
  • [2] Automatic Hypernasality Detection in Cleft Palate Speech Using CNN
    Wang, Xiyue
    Tang, Ming
    Yang, Sen
    Yin, Heng
    Huang, Hua
    He, Ling
    CIRCUITS SYSTEMS AND SIGNAL PROCESSING, 2019, 38 (08) : 3521 - 3547
  • [3] Automatic Evaluation of Hypernasality Based on a Cleft Palate Speech Database
    He, Ling
    Zhang, Jing
    Liu, Qi
    Yin, Heng
    Lech, Margaret
    Huang, Yunzhi
    JOURNAL OF MEDICAL SYSTEMS, 2015, 39 (05) : 1 - 7
  • [4] Automatic Evaluation of Hypernasality and Consonant Misarticulation in Cleft Palate Speech
    He, Ling
    Zhang, Jing
    Liu, Qi
    Yin, Heng
    Lech, Margaret
    IEEE SIGNAL PROCESSING LETTERS, 2014, 21 (10) : 1298 - 1301
  • [5] Estimation of Hypernasality Scores from Cleft Lip and Palate Speech
    Vikram, C. M.
    Tripathi, Ayush
    Kalita, Sishir
    Prasanna, S. R. Mahadeva
    19TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2018), VOLS 1-6: SPEECH RESEARCH FOR EMERGING MARKETS IN MULTILINGUAL SOCIETIES, 2018, : 1701 - 1705
  • [6] Variational mode decomposition based features for detection of hypernasality in cleft palate speech
    Sireesha, Kilaru
    Dubey, Akhilesh Kumar
    Govind, D.
    Samudravijaya, K.
    Gangashetty, Suryakanth, V
    BIOMEDICAL SIGNAL PROCESSING AND CONTROL, 2024, 97
  • [7] Automatic Evaluation of Hypernasality Based on a Cleft Palate Speech Database
    Ling He
    Jing Zhang
    Qi Liu
    Heng Yin
    Margaret Lech
    Yunzhi Huang
    Journal of Medical Systems, 2015, 39
  • [8] Automatic Hypernasality Detection in Cleft Palate Speech Using CNN
    Xiyue Wang
    Ming Tang
    Sen Yang
    Heng Yin
    Hua Huang
    Ling He
    Circuits, Systems, and Signal Processing, 2019, 38 : 3521 - 3547
  • [9] Automatic hypernasality grade assessment in cleft palate speech based on the spectral envelope method
    Zhang, Jing
    Yang, Sen
    Wang, Xiyue
    Tang, Ming
    Yin, Heng
    He, Ling
    BIOMEDICAL ENGINEERING-BIOMEDIZINISCHE TECHNIK, 2020, 65 (01): : 73 - 86
  • [10] Spectral Analysis Based Objective Assessment of Hypernasality in Cleft Palate Speech: A Review
    Dubey, Akhilesh Kumar
    Prasanna, S. R. Mahadeva
    Dandapat, S.
    CIRCUITS SYSTEMS AND SIGNAL PROCESSING, 2025, : 4133 - 4166