Multi-task Learning for Speaker Verification and Voice Trigger Detection
This addresses the challenge of interdependent speech tasks for applications like voice assistants, but it is incremental as it combines existing methods without major breakthroughs.
The study tackled the problem of training a single network to jointly perform speech transcription and speaker recognition, showing that it achieved accuracies at least as good as baseline models for both tasks while using the same number of parameters.
Automatic speech transcription and speaker recognition are usually treated as separate tasks even though they are interdependent. In this study, we investigate training a single network to perform both tasks jointly. We train the network in a supervised multi-task learning setup, where the speech transcription branch of the network is trained to minimise a phonetic connectionist temporal classification (CTC) loss while the speaker recognition branch of the network is trained to label the input sequence with the correct label for the speaker. We present a large-scale empirical study where the model is trained using several thousand hours of labelled training data for each task. We evaluate the speech transcription branch of the network on a voice trigger detection task while the speaker recognition branch is evaluated on a speaker verification task. Results demonstrate that the network is able to encode both phonetic \emph{and} speaker information in its learnt representations while yielding accuracies at least as good as the baseline models for each task, with the same number of parameters as the independent models.