Personalizing Keyword Spotting with Speaker Information
This addresses the challenge of personalizing keyword spotting for real-world applications with diverse populations, though it is incremental as it builds on existing FiLM methods.
The paper tackled the problem of keyword spotting systems struggling with diverse accents and age groups by integrating speaker information using Feature-wise Linear Modulation (FiLM), resulting in a substantial improvement in keyword detection accuracy, especially for underrepresented groups, with only a 1% increase in parameters and minimal impact on latency.
Keyword spotting systems often struggle to generalize to a diverse population with various accents and age groups. To address this challenge, we propose a novel approach that integrates speaker information into keyword spotting using Feature-wise Linear Modulation (FiLM), a recent method for learning from multiple sources of information. We explore both Text-Dependent and Text-Independent speaker recognition systems to extract speaker information, and we experiment on extracting this information from both the input audio and pre-enrolled user audio. We evaluate our systems on a diverse dataset and achieve a substantial improvement in keyword detection accuracy, particularly among underrepresented speaker groups. Moreover, our proposed approach only requires a small 1% increase in the number of parameters, with a minimum impact on latency and computational cost, which makes it a practical solution for real-world applications.