Study of Hidden Markov Models and Neural Networks for Robust Isolated Word Recognition


Keywords neural networks, noise robustness, speech recognition

Start Date: 01 October 1992 / Duration: 36 months / Status: ongoing

[ contact / participants ]

the development and assessment of neural network techniques for improving the robustness of medium vocabulary (50-100 words), speaker-independent, isolated word recognisers for telephone transmission quality speech.

Objectives and Approach

The dominant technology is Hidden Markov Models (HMMs) but this has significant limitations, some of which could be alleviated by the judicious use of artificial neural networks (ANNs) or hybrid combinations of both techniques.

Direct comparisons of ANN-based, HMM-based, and hybrid ANN/HMM techniques for speech recognition will be made. The developments will be integrated and validated in the context of a telephone application including speech recognition capabilities.

A number of prototypes have been demonstrated on low cost commodity systems.

The telephone application developed within the project will be the basis for product development by Tedas.

Further information about HIMARNNET is available from the HIMARNNET home page <URL:>.


Henri Leich
Faculté Polytechnique de Mons
Blvd Dolez 31
B-7000 Mons
tel: +32 6537 4128
fax: +32 6537 4300


Faculté Polytechnique de Mons [B]
ASCOM Holding Ltd [CH]
École Polytechnique Federale de Lausanne [CH]
Lernout & Hauspie Speech Products [B]
Tedas Gesellschaft für Technische Datenverarbeitung mbH [D]

HPCN synopses home page HPCN acronym index HPCN number index
All synopses home page all acronyms index all numbers index

HIMARNNET - 6488, December 1993

please address enquiries to the ESPRIT Information Desk

html version of synopsis by Nick Cook