Deep learning architectures for estimating breathing signal and respiratory parameters from speech recordings

Materials-Driven Soft Wearable Bioelectronics for Connected Healthcare

In the era of Internet-of-things, many things can stay connected; however, biological systems, including those necessary for human health, remain unable to stay connected to the global Internet due to the lack of soft conformal biosensors. The fundamental challenge lies in the fact that electronics and biology are distinct and incompatible, as they are based on different materials via different functioning principles. In particular, the human body is soft and curvilinear, yet electronics are typically rigid and planar. Recent advances in materials and materials design have generated tremendous opportunities to design soft wearable bioelectronics, which may bridge the gap, enabling the ultimate dream of connected healthcare for anyone, anytime, and anywhere. We begin with a review of the historical development of healthcare, indicating the significant trend of connected healthcare. This is followed by the focal point of discussion about new materials and materials design, particularly low-dimensional nanomaterials. We summarize material types and their attributes for designing soft bioelectronic sensors; we also cover their synthesis and fabrication methods, including top-down, bottom-up, and their combined approaches. Next, we discuss the wearable energy challenges and progress made to date. In addition to front-end wearable devices, we also describe back-end machine learning algorithms, artificial intelligence, telecommunication, and software. Afterward, we describe the integration of soft wearable bioelectronic systems which have been applied in various testbeds in real-world settings, including laboratories that are preclinical and clinical environments. Finally, we narrate the remaining challenges and opportunities in conjunction with our perspectives.

Automatic Breathing Pattern Analysis from Reading-Speech Signals

As the speech production mechanism is related to the breathing process, speech signals and breathing patterns impact each other. Breathing patterns are the physiological signals which help in understanding the psychological, physiological and cognitive states of an individual. Capturing such patterns relies on the availability of equipment such as respiratory belts, which are costly and uncomfortable to wear for long duration. In this paper, we attempt to extract the breathing patterns from speech signals, which are easily available and can be recorded using a smartphone's microphone. In the presented work, simultaneous speech and breath signals are captured from 100 Indians of the age group 20 to 25 years while they read a phonetically balanced passage in English language. We have identified five distinct breathing templates; following two broad speech-breath categories, exhibited by the speakers while they read the same passage. For one of the two categories, the time domain features with regression network can extract the breathing patterns from speech with a Pearson correlation coefficient of 0.70. By computational modelling, we distinguish these two breathing categories from speech with a classification accuracy of 79%.

Immediate Effects of Mindfulness Meditation on the Voice

OBJECTIVES

The benefits of mindfulness meditation are well documented. This study evaluated the immediate effects of mindfulness meditation (MM) on the voice and voice user.

STUDY DESIGN

Prospective experimental study.

METHODS

Participants: 19 vocally healthy (VH) individuals, and 26 individuals with common voice disorders (CVD; benign lesions and hyperfunctional muscle tension) deemed stimulable for voice therapy. Exclusionary criteria: prior training or regular meditation practice. Participants recorded speech samples before and after a 11.5-minute prerecorded session of MM.

PRIMARY OUTCOMES

phonatory aerodynamics and participants' self-reported experience of voice.

SECONDARY OUTCOMES

self-reported anxiety, vocal acoustics, speech breathing patterns, and auditory-perceptual outcomes. Baseline self-reported measures of voice (Voice Handicap Index-10 - VHI-10), breathing (Dyspnea Index - DI), stress (Perceived Stress Scale - PSS), and trait mindfulness (Cognitive and Mindfulness Scale - Revised, CAMS-R, Five Facet Mindfulness Questionnaire - FFMQ) were compared between groups.

RESULTS

At baseline, CVD had significantly higher VHI-10 (P< 0.001) and DI (P= 0.0014), and lower trait mindfulness (CAMS-R, P= 0.02). No difference between groups for PSS or FFMQ. Changes postMM: decreased CPP for all-voiced sentences for VH (P= 0.003), decreased mean SPL (P= 0.012) on sustained vowel for VH, increased mean phonatory airflow during sustained vowel for CVD (P = 0.012). VH demonstrated a decrease in CPP on the all-voice sentence, and CVD demonstrated an increase, resulting in a significant between group difference (P= 0.013). Participants reported improvements in voice, emotional and physical states. State anxiety decreased for both groups (= < 0.001). No other objective outcomes reached significance.

CONCLUSIONS

After a brief MM, participants experienced improvement in physical, emotional, and cognitive states, and in their perceptions of their voice. Results indicate that a brief, single session of MM may be beneficial for some, but not sufficient to override habitual voice and speech patterns. Given the benefits of MM, future work should evaluate MM in a standard voice therapy protocol.