An ingestion monitoring system comprises one or more acoustic transducers, an analogue to digital converter, and a processor arranged to process the digital signal to detect an ingestion event. The system can be a wearable device, like a wristband, a necklace, an ankle band or an ear hook, and the processing circuitry can be attached to the device, or be remote and wirelessly connected. The acoustic signals can be pre-processed using a low-pass filter to detect low frequencies and taking a short term Fourier transform to observe high frequency components. Other sensors, like motion, heart rate, photoplethysmographic, temperature and satellite navigation (GPS) sensors may be included in the system, and a combination of the data from all, or a subset of sensors may be analysed using statistical methods, like Gaussian mixture models or Bayesian inference to identify certain user actions (like eating or drinking) from the data, or hidden Markov models can be applied on the data to identify certain patterns. Also disclosed is a computer program product that processes a digital signal to detect an ingestion event.
