Synth4Kws: Synthesized Speech for User Defined Keyword Spotting in Low Resource Environments
This addresses the problem of data scarcity for developers of custom keyword spotting systems, offering a practical solution that is incremental but effective.
The paper tackles the challenge of expensive data collection for custom keyword spotting by introducing Synth4Kws, a framework that uses text-to-speech synthesized data to improve model performance; results show that with optimal TTS data, error rates can be reduced by 30.1% and AUC improved by 46.7% in low-resource settings.
One of the challenges in developing a high quality custom keyword spotting (KWS) model is the lengthy and expensive process of collecting training data covering a wide range of languages, phrases and speaking styles. We introduce Synth4Kws - a framework to leverage Text to Speech (TTS) synthesized data for custom KWS in different resource settings. With no real data, we found increasing TTS phrase diversity and utterance sampling monotonically improves model performance, as evaluated by EER and AUC metrics over 11k utterances of the speech command dataset. In low resource settings, with 50k real utterances as a baseline, we found using optimal amounts of TTS data can improve EER by 30.1% and AUC by 46.7%. Furthermore, we mix TTS data with varying amounts of real data and interpolate the real data needed to achieve various quality targets. Our experiments are based on English and single word utterances but the findings generalize to i18n languages and other keyword types.