What shall we do with an hour of data? Speech recognition for the un- and under-served languages of Common Voice
This work addresses the problem of speech recognition for un- and under-served languages, but it is incremental as it applies existing methods to new data.
The researchers tackled the challenge of creating deployable speech recognition models for 31 under-served languages using limited data from the Common Voice project, achieving specific accuracy results on official testing sets.
This technical report describes the methods and results of a three-week sprint to produce deployable speech recognition models for 31 under-served languages of the Common Voice project. We outline the preprocessing steps, hyperparameter selection, and resulting accuracy on official testing sets. In addition to this we evaluate the models on multiple tasks: closed-vocabulary speech recognition, pre-transcription, forced alignment, and key-word spotting. The following experiments use Coqui STT, a toolkit for training and deployment of neural Speech-to-Text models.