Initial evaluation of a preselection module for a flexible large vocabulary speech recognition system in telephone environment

Abstract

We are improving a flexible, large vocabulary, speaker independent, isolated-word recognition system in a telephone environment, originally designed as an integrated system doing all the recognition process in one step. We have transformed it, by adopting the hypothesis-verification paradigm.In this paper, we will describe the architecture and results of the hypothesis subsystem. We will show the system evolution and the modifications adopted to face such a difficult task, achieving significant improvements using automatically clustered phonemelike units, semi-continuous HMMs, and multiple models per unit. Also, system behavior for vocabulary dependent and independent tasks and vocabularies up to 10000 words will be tested.

Recent work on a preselection module for a flexible large vocabulary speech recognition system in telephone environment

“…The hypothesis module divides the recognition process in two: the first one generates a phonetic string, which is taken by a lexical access module to give a list of candidate words to the verification stage. A detailed description can be found in [6].…”

“…

At ICSLP'96 we presented a flexible, large vocabulary, speaker independent, isolated-word preselection system in a telephone environment, using a two stage, bottom-up strategy [6]. We achieved reasonable performance in large and very large vocabulary tasks, ranging from 1200 to 10000 words.In this paper, we describe recent studies we have carried out on the system, aimed at two directions: handling of non speech sounds in the speech signal (we consider lips, respiration and click noises); and making the preselection lists dynamic in length, to reduce computational load, in the average.

…”

“…At ICSLP'96 we presented a flexible, large vocabulary, speaker independent, isolated-word preselection system in a telephone environment, using a two stage, bottom-up strategy [6]. We achieved reasonable performance in large and very large vocabulary tasks, ranging from 1200 to 10000 words.…”

Mejora de servicios por teléfono con reconocimiento de habla. Nueva generación de servidores vocales interactivos

“…Como resultado del reconocimiento se ofrece una lista ordenada de candidatos. El reconocimiento se realiza en dos etapas: hipótesis (preselección) y verificación (reconocimiento más detallado) (Macías-Guarasa et al, 1996b;Ferreiros et al, 1998;Macías-Guarasa et al, 2000a;Córdoba et al, 2001).…”