Inclusion Criteria：

1. Breathing is steady at rest: no apparent dyspnea, chest tightness, or orthopnea, etc.
2. Heart rate, blood pressure, and SpO₂ are all stable overall.SpO₂ reduction from baseline < 5%.
3. No frequent need for emergency sputum suction or indications of sputum blockage while wearing it.

Once patient conditions stabilized, they were all transitioned to the decannulation process established by our department in 2018, which utilizes a speaking valve instead of capping.The initial duration for the patient to wear the speaking valve is 30 minutes.If the patient can tolerate wearing the speaking valve for 30 minutes, they fulfill the inclusion criteria.The monitoring period for wearable devices is from 1 hour before the initial use of the speaking valve to 1 hour after its removal.Monitored for a total of 2 hours and 30 minutes.

Physiological parameters recorded prior to wearing the speaking valve served as the baseline.Compare the physiological parameters while wearing the speaking valve and after its removal to investigate the clinical benefits that the speaking valve can provide.To promote the application of speaking valves in future clinical work more effectively

Previous studies have been concerned that speaking valve placement increases upper airway resistance, thereby inducing dyspnea and hypoxemia in patients. Therefore, we used continuous wearable physiological monitoring devices to observe changes in oxygen saturation before, during, and after the training.

The rate of change (percentage per minute) in asymmetric cardiovascular responses (e.g., heart rate variability indices) and delayed autonomic recovery will be calculated. Physiological data (heart rate, HRV metrics) are continuously collected before, during, and after speaking valve placement. The primary reported value is the slope of change over time, expressed as a percentage relative to baseline, aggregated across all participants using mixed-effects modeling to produce a single mean rate of change

The rate of change in respiratory phase variation (e.g., changes in inspiratory/expiratory time proportion) during speaking valve training. Respiratory parameters are continuously recorded. The outcome is the calculated rate of change (percentage per minute) from baseline to the active training period and to the recovery period. A single aggregated value (mean slope) will be derived from all participants using linear mixed models.