Improving Daily Routine Recognition in Hearing Aids Using Sequence Learning

Abstract

This work focuses on sequence learning to improve the daily routine recognition in hearing aids (HA), where the goal is to personalize the device configuration for each user. We apply the sequence methods on two large real-world data sets. One publicly available set contains the acceleration (ACC) data of one person, Huynh, over seven working days, whereas our set includes the real life of seven subjects over 104 days with ACC and audio data of a HA. For both sets, we design statistical features to represent the recurring routine behavior well. In our comprehensive simulations, we analyze several sequence classifiers learning the temporal relationships of high-level activities. The multi-layer perceptron (MLP) and random forest (RF) as an observation model for the hidden Markov model (HMM) show the best F-measure performance of 85.3% and 91.6% on our set and the Huynh set, respectively. In particular, the MLP-HMM combination strongly improves on both sets compared to the non-sequence classifier MLP by 6.7% and 10.2%. Within the segment error analysis, we show that the sequence classifiers improve the temporal prediction stability by a reduction of insertion errors. Thus, the improved sequence classification helps the user to better address his condition due to preferred HA settings.

A Review of Machine Learning Approaches for the Personalization of Amplification in Hearing Aids

“…In [ 46 ], daily routines were recognized using a sequence-learning model that involved feature representation and using a Long Short-Term Memory (LSTM) network together with a Hidden Markov Model (HMM) for sequence behavior learning. The features were derived from two accelerometer (ACC) sensors and audio data, and a statistical representation was developed on both activity primitive and routine levels.…”

Predicting and Explaining Hearing Aid Usage Using Encoder-Decoder with Attention Mechanism and SHAP

“…Among these models, recurrent-based Deep Learning (DL) models such as Long Short-Term Memory (LSTM) (Hochreiter & Schmidhuber, 1997) are particularly popular due to their ability to handle the sequential properties found in many HLrelated datasets, as well as capable of dealing both long-term and short-term time series (Preeti et al, 2019). LSTM has been applied to a number of HL-related fields, including speech assessment (Chiang et al, 2021), speech enhancements (Garg, 2022;Zhang et al, 2019), as well as daily routine recognition using acceleration and audio data from the HAids (Kuebert et al, 2021).…”