Enhancing spontaneous speech recognition with BLSTM features

Martin Wöllmer, Björn Schuller

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

This paper introduces a novel context-sensitive feature extraction approach for spontaneous speech recognition. As bidirectional Long Short-Term Memory (BLSTM) networks are known to enable improved phoneme recognition accuracies by incorporating long-range contextual information into speech decoding, we integrate the BLSTM principle into a Tandem front-end for probabilistic feature extraction. Unlike previously proposed approaches which exploit BLSTM modeling by generating a discrete phoneme prediction feature, our feature extractor merges continuous high-level probabilistic BLSTM features with low-level features. Evaluations on challenging spontaneous, conversational speech recognition tasks show that this concept prevails over recently published architectures for feature-level context modeling.

Original languageEnglish
Title of host publicationAdvances in Nonlinear Speech Processing - 5th International Conference on Nonlinear Speech Processing, NOLISP 2011, Proceedings
Pages17-24
Number of pages8
DOIs
StatePublished - 2011
Event5th International Conference on Nonlinear Speech Processing, NOLISP 2011 - Las Palmas de Gran Canaria, Spain
Duration: 7 Nov 20119 Nov 2011

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7015 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference5th International Conference on Nonlinear Speech Processing, NOLISP 2011
Country/TerritorySpain
CityLas Palmas de Gran Canaria
Period7/11/119/11/11

Keywords

  • bidirectional neural networks
  • context modeling
  • probabilistic features
  • speech recognition

Fingerprint

Dive into the research topics of 'Enhancing spontaneous speech recognition with BLSTM features'. Together they form a unique fingerprint.

Cite this