Product detail

Tensorflow implementation of speaker recognition with x-vector topology

ZEINALI, H. BURGET, L. ROHDIN, J. STAFYLAKIS, T. ČERNOCKÝ, J.

Product type

software

Abstract

This is a Tensorflow implementation of x-vector topology (speaker embedding). It uses Kaldi toolkit for data processing. We train the model using Tensorflow and also extract speaker embeddings (x-vectors) using it. This allow to train or retrain the system to the particular customer specific domain or provides the ability to modify the topology or training schema to achieve better performance for the specific domain.  This software is a result of Czech Ministry of Interior project  "Dolování infoRmAcí z řeči Pořízené vzdÁlenými miKrofony - DRAPÁK", No. VI20152020025 (https://www.fit.vut.cz/research/project/1009/)

Keywords

Speaker recognition, speaker embedding, DNN, x-vectors, retraining

Create date

12. 5. 2019

Location

https://github.com/BUTSpeechFIT/x-vector-kaldi-tf

Possibilities of use

Využití výsledku jiným subjektem je možné bez nabytí licence (výsledek není licencován)

Licence fee

Poskytovatel licence na výsledek nepožaduje licenční poplatek

www

https://github.com/BUTSpeechFIT/x-vector-kaldi-tf

Documents

zeinali_icassp2019_0006141.pdf