Analysis of X-Vectors for Low-Resource Speech Recognition

Martin Karafiát; Karel Veselý; Jan Honza Cernocký; Ján Profant; Jirí Nytra; Miroslav Hlavácek; Tomás Pavlícek

ICASSP 2021accepted0 citations

Analysis of X-Vectors for Low-Resource Speech Recognition

Martin Karafiát, Karel Veselý, Jan Honza Cernocký, Ján Profant, Jirí Nytra, Miroslav Hlavácek, Tomás Pavlícek

Abstract

The paper presents a study of usability of x-vectors for adaptation of automatic speech recognition (ASR) systems. X-vectors are Neural Network (NN)-based speaker embeddings recently proposed in speaker recognition (SR). They quickly replaced common i-vectors and became new state-of-the-art technique. Here, the same approach is adopted for ASR with the hope of similar outcome. All experiments were done on ASR for the latest IARPA MATERIAL evaluation running on Pashto language. Over 1% absolute improvement was observed with x-vectors over traditional i-vectors, even when the x-vector extractor was not trained on target Pashto data.

BibTeX

@inproceedings{icassp2021_analysisofxvecto,
  title = {Analysis of X-Vectors for Low-Resource Speech Recognition},
  author = {Martin Karafiát and Karel Veselý and Jan Honza Cernocký and Ján Profant and Jirí Nytra and Miroslav Hlavácek and Tomás Pavlícek},
  booktitle = {ICASSP 2021},
  year = {2021}
}