Publication detail

Semi-Supervised Bootstrapping Approach For Neural Network Feature Extractor Training

GRÉZL, F. KARAFIÁT, M.

Original Title

Semi-Supervised Bootstrapping Approach For Neural Network Feature Extractor Training

Type

article in a collection out of WoS and Scopus

Language

English

Original Abstract

This paper presents bootstrapping approach for training the Bottle-Neck neural network feature extractor which provides features  for subsequent GMM-HMM recognizer. One can use this recognizer to automatically transcribe the unsupervised data and assign the confidence of the transcription. Based on the confidence, segments are selected and mixed with supervised data and new NNs are trained. The automatic transcription can recover 40-55% in comparison to manually transcribed data. This is 3 to 5% absolute improvement over NN trained on supervised data only. Using 70-85% of automatically transcribed segments with the highest confidence was found optimal to achieve this result. Dropping the rest of the data prevents training on low quality transcripts.

Keywords

Semi-supervised training, bootstrapping, bottle-neck features

Authors

GRÉZL, F.; KARAFIÁT, M.

RIV year

2013

Released

8. 12. 2013

Publisher

IEEE Signal Processing Society

Location

Olomouc

ISBN

978-1-4799-2755-5

Book

Proceedings of ASRU 2013

Pages from

470

Pages to

475

Pages count

6

URL

BibTex

@inproceedings{BUT105972,
  author="František {Grézl} and Martin {Karafiát}",
  title="Semi-Supervised Bootstrapping Approach For Neural Network Feature Extractor Training",
  booktitle="Proceedings of ASRU 2013",
  year="2013",
  pages="470--475",
  publisher="IEEE Signal Processing Society",
  address="Olomouc",
  isbn="978-1-4799-2755-5",
  url="http://www.fit.vutbr.cz/research/groups/speech/publi/2013/grezl_asru2013_0000470.pdf"
}