Adapting to Noise in Speech Obfuscation by Audio Profiling Using Generative Models for Passive Health Monitoring

Passive health monitoring has been introduced as a solution for continuous diagnosis and tracking of subjects’ condition with minimal effort. This is partially achieved by the technology of passive audio recording although it poses major audio privacy issues for subjects. Existing methods are limited to controlled recording environments and their prediction is significantly influenced by background noises. Meanwhile, they are too compute-intensive to be continuously running on smart phones. In this paper, we implement an efficient and robust audio privacy preserving method that profiles the background audio to focus only on audio activities detected during recording for performance improvement, and to adapt to the noise for more accurate speech segmentation. We analyze the performance of our method using audio data collected by a smart watch in lab noisy settings. Our obfuscation results show a low false positive rate of 20% with a 92% true positive rate by adapting to the recording noise level. We also reduced model memory footprint and execution time of the method on a smart phone by 75% and 62% to enable continuous speech obfuscation.

[1]  Mani B. Srivastava,et al.  Privacy risks emerging from the adoption of innocuous wearable sensors in the mobile environment , 2011, CHI.

[2]  Pete Warden,et al.  Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition , 2018, ArXiv.

[3]  Ke Wang,et al.  AI Benchmark: Running Deep Neural Networks on Android Smartphones , 2018, ECCV Workshops.

[4]  Ming Zeng,et al.  Sound Shredding: Privacy Preserved Audio Sensing , 2015, HotMobile.

[5]  Youngduck Choi,et al.  Early Detection of Diabetes from Health Claims , 2013 .

[6]  Jilong Kuang,et al.  Assessment of Chronic Pulmonary Disease Patients Using Biomarkers from Natural Speech Recorded by Mobile Devices , 2019, 2019 IEEE 16th International Conference on Wearable and Implantable Body Sensor Networks (BSN).

[7]  Upkar Varshney,et al.  Pervasive Healthcare and Wireless Health Monitoring , 2007, Mob. Networks Appl..

[8]  Jilong Kuang,et al.  Assessing Severity of Pulmonary Obstruction from Respiration Phase-Based Wheeze-Sensing Using Mobile Sensors , 2020, CHI.

[9]  Xiaohui Liang,et al.  ESPAC: Enabling Security and Patient-centric Access Control for eHealth in cloud computing , 2011, Int. J. Secur. Networks.

[10]  Yuan Yu,et al.  TensorFlow: A system for large-scale machine learning , 2016, OSDI.

[11]  Vedran Bilas,et al.  Low-Power Wearable Respiratory Sound Sensing , 2014, Sensors.

[12]  Laurent Girin,et al.  Deep neural networks for automatic detection of screams and shouted speech in subway trains , 2016, 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[13]  Jilong Kuang,et al.  A Generative Model for Speech Segmentation and Obfuscation for Remote Health Monitoring , 2019, 2019 IEEE 16th International Conference on Wearable and Implantable Body Sensor Networks (BSN).

[14]  Jilong Kuang,et al.  Configurable Pulmonary-Tuned Privacy Preservation Algorithm for Mobile Devices , 2018, 2018 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[15]  Chris Donahue,et al.  Synthesizing Audio with Generative Adversarial Networks , 2018, ArXiv.

[16]  John Adcock,et al.  Audio privacy: reducing speech intelligibility while preserving environmental sounds , 2008, ACM Multimedia.

[17]  Justin Salamon,et al.  Deep Convolutional Neural Networks and Data Augmentation for Environmental Sound Classification , 2016, IEEE Signal Processing Letters.

[18]  Lipika Dey,et al.  A k-mean clustering algorithm for mixed numeric and categorical data , 2007, Data Knowl. Eng..

[19]  Jilong Kuang,et al.  Extraction of Voice Parameters from Continuous Running Speech for Pulmonary Disease Monitoring , 2019, 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).

[20]  Audie A Atienza,et al.  Mobile health technology evaluation: the mHealth evidence workshop. , 2013, American journal of preventive medicine.

[21]  Xiaodong Lin,et al.  Sage: a strong privacy-preserving scheme against global eavesdropping for ehealth systems , 2009, IEEE Journal on Selected Areas in Communications.

[22]  Jilong Kuang,et al.  Private Audio-Based Cough Sensing for In-Home Pulmonary Assessment Using Mobile Devices , 2018, BODYNETS.

[23]  Jilong Kuang,et al.  mLung++: automated characterization of abnormal lung sounds in pulmonary patients using multimodal mobile sensors , 2019, UbiComp/ISWC Adjunct.

[24]  A. Lymberis,et al.  Smart wearables for remote health monitoring, from prevention to rehabilitation: current R&D, future challenges , 2003, 4th International IEEE EMBS Special Topic Conference on Information Technology Applications in Biomedicine, 2003..

[25]  Vitaly Shmatikov,et al.  Myths and fallacies of "Personally Identifiable Information" , 2010, Commun. ACM.

[26]  Tomi Kinnunen,et al.  A practical, self-adaptive voice activity detector for speaker verification with noisy telephone and microphone data , 2013, 2013 IEEE International Conference on Acoustics, Speech and Signal Processing.