Abstract
This paper is focused on the finetuning of acoustic models for speaker adap-tation based on a given gender. We pretrained the Transformer baseline model on Librispeech-960 and conduct experiments with finetuning on the gender-specific test subsets. Our approach leads to 5% lower word error rate on the male subset if the layers in the encoder and decoder are not frozen, but the tuning is started from the last checkpoints. Moreover, we adapted our general model on the full L2 Arctic dataset of accented speech and finetuned it for particular speakers and male and female genders separately. The models trained on the gender subsets obtained 1-2% higher accuracy when compared to the model tuned on the whole L2 Arctic dataset. Finally, we tested the concatenation of the pretrained x-vector voice embeddings and embeddings from conventional encoder, but its gain in accuracy is not significant.
Abstract (translated)
URL
https://arxiv.org/abs/2010.04224
