Predicting Heart Rate from Speech Signals

Heart disease is known to be one of the main cause of death worldwide. In the United States, the disease is the leading cause of death for people of most ethnicity. In Indonesia, the disease contributes 17% of the total mortality. It is also well-known that a regular or continuous monitoring the heart rate is important to minimize the fatality. At the current technological state, the monitoring method is expensive and often requires wearable sensor, which is inconvenient. This study intends to establish a relationship between the heart rate and speech signals empirically by using a regression and correlation analysis. The speech signals are analyzed through a frequency analysis using fast Fourier transformation. The results suggest that vowel ā€˜eā€™ speech signal has a strong correlation to the heart rate.

Fergyanto E Gunawan, Dr Eng and Mellisa Pratiwi