SpeakerStew: Scaling to Many Languages with a Triaged Multilingual Text-Dependent and Text-Independent Speaker Verification System
This addresses the problem of scaling speaker verification to many languages for smart speaker devices, with incremental improvements in efficiency.
The paper tackles speaker verification across 46 languages by developing a hybrid system that pools multilingual training data for generalization and uses a triage mechanism between text-dependent and text-independent models to reduce computational costs. Results show training on multiple languages reduces computational requirements by an order of magnitude and the triage framework cuts calls to the expensive text-independent system by 73% while maintaining performance.
In this paper, we describe SpeakerStew - a hybrid system to perform speaker verification on 46 languages. Two core ideas were explored in this system: (1) Pooling training data of different languages together for multilingual generalization and reducing development cycles; (2) A novel triage mechanism between text-dependent and text-independent models to reduce runtime cost and expected latency. To the best of our knowledge, this is the first study of speaker verification systems at the scale of 46 languages. The problem is framed from the perspective of using a smart speaker device with interactions consisting of a wake-up keyword (text-dependent) followed by a speech query (text-independent). Experimental evidence suggests that training on multiple languages can generalize to unseen varieties while maintaining performance on seen varieties. We also found that it can reduce computational requirements for training models by an order of magnitude. Furthermore, during model inference on English data, we observe that leveraging a triage framework can reduce the number of calls to the more computationally expensive text-independent system by 73% (and reduce latency by 59%) while maintaining an EER no worse than the text-independent setup.