Uncovering Voice Misuse Using Symbolic Mismatch

Voice disorders affect an estimated 14 million working-aged Americans, and many more worldwide. We present the first large scale study of vocal misuse based on long-term ambulatory data collected by an accelerometer placed on the neck. We investigate an unsupervised data mining approach to uncovering latent information about voice misuse. We segment signals from over 253 days of data from 22 subjects into over a hundred million single glottal pulses (closures of the vocal folds), cluster segments into symbols, and use symbolic mismatch to uncover differences between patients and matched controls, and between patients pre- and post-treatment. Our results show significant behavioral differences between patients and controls, as well as between some pre- and post-treatment patients. Our proposed approach provides an objective basis for helping diagnose behavioral voice disorders, and is a first step towards a more data-driven understanding of the impact of voice therapy.

[1]  Eamonn J. Keogh,et al.  Experimental comparison of representation methods and distance measures for time series data , 2010, Data Mining and Knowledge Discovery.

[2]  Eamonn J. Keogh,et al.  A symbolic representation of time series, with implications for streaming algorithms , 2003, DMKD '03.

[3]  J. Laver.  The phonetic description of voice quality , 1980 .

[4]  Susan Miller,et al.  Incidence of supraglottic activity in males and females: a preliminary report. , 2003, Journal of voice : official journal of the Voice Foundation.

[5]  H. Kingma,et al.  Vocal load as measured by the voice accumulator. , 1995, Folia phoniatrica et logopaedica : official organ of the International Association of Logopedics and Phoniatrics.

[6]  M. Hsiung,et al.  The Characteristic Features of Muscle Tension Dysphonia before and after Surgery in Benign Lesions of the Vocal Fold , 2004, ORL.

[7]  Eamonn J. Keogh,et al.  Mining motifs in massive time series databases , 2002, 2002 IEEE International Conference on Data Mining, 2002. Proceedings..

[8]  Zeeshan Syed,et al.  Clustering and Symbolic Analysis of Cardiovascular Signals: Discovery and Visualization of Medically Relevant Patterns in Long-Term Data Using Limited Prior Knowledge , 2007, EURASIP J. Adv. Signal Process..

[9]  D. Mehta,et al.  Evidence-based clinical voice assessment: a systematic review. , 2013, American journal of speech-language pathology.

[10]  Robert E. Hillman,et al.  Mobile Voice Health Monitoring Using a Wearable Accelerometer Sensor and a Smartphone Platform , 2012, IEEE Transactions on Biomedical Engineering.

[11]  Suchi Saria,et al.  Discovering shared and individual latent structure in multiple time series , 2010, ArXiv.

[12]  P H Dejonckere,et al.  Neurovegetative symptoms and complaints before and after voice therapy for nonorganic habitual dysphonia. , 2008, Journal of voice : official journal of the Voice Foundation.

[13]  E Vilkman,et al.  Relationship between subjective voice complaints and acoustic parameters in female teachers' voices. , 1999, Journal of voice : official journal of the Voice Foundation.

[14]  Edwin Lughofer,et al.  Extensions of vector quantization for incremental clustering , 2008, Pattern Recognit..

[15]  Marzyeh Ghassemi,et al.  Learning to Detect Vocal Hyperfunction From Ambulatory Neck-Surface Acceleration Features: Initial Results for Vocal Fold Nodules , 2014, IEEE Transactions on Biomedical Engineering.

[16]  Zeeshan Syed,et al.  Unsupervised Similarity-Based Risk Stratification for Cardiovascular Events Using Long-Term Time-Series Data , 2011, J. Mach. Learn. Res..

[17]  S. Gray,et al.  Voice Disorders in the General Population: Prevalence, Risk Factors, and Occupational Impact , 2005, The Laryngoscope.

[18]  H. K. Schutte,et al.  Anterior-posterior and medial compression of the supraglottis: signs of nonorganic dysphonia or normal postures? , 2003, Journal of voice : official journal of the Voice Foundation.

[19]  A. Gatherer.  Clinical , 1997 .

[20]  Q Li,et al.  Dynamic time warping and machine learning for signal quality assessment of pulsatile signals , 2012, Physiological measurement.