Speech recognition using sub-word neural tree network models and multiple classifier fusion

Manish Sharma, Richard Mammone

Research output: Contribution to journalConference articlepeer-review

Abstract

A new neural tree network (NTN) -based speech recognition system is presented. In the sub-word unit-based system, the NTNs model sub-word speech segments. Durational probability is associated with each sub-word NTN. An iterative algorithm is proposed for training the sub-word NTNs. The sub-word NTN models, as well as the sub-word segment boundaries within a vocabulary word, are re-estimated. Thus, this paradigm can be argued to represent a class of discriminatory training-based, homogeneous, sub-word unit-based, speech recognition systems. Hence, the results reported can be generalized to other similar systems.

Original languageEnglish (US)
Pages (from-to)3323-3326
Number of pages4
JournalICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume5
StatePublished - 1995
EventProceedings of the 1995 20th International Conference on Acoustics, Speech, and Signal Processing. Part 2 (of 5) - Detroit, MI, USA
Duration: May 9 1995May 12 1995

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Speech recognition using sub-word neural tree network models and multiple classifier fusion'. Together they form a unique fingerprint.

Cite this