Paper: lex4all: A language-independent tool for building and evaluating pronunciation lexicons for small-vocabulary speech recognition

ACL ID P14-5019
Title lex4all: A language-independent tool for building and evaluating pronunciation lexicons for small-vocabulary speech recognition
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2014
Authors

This paper describes lex4all, an open- source PC application for the generation and evaluation of pronunciation lexicons in any language. With just a few minutes of recorded audio and no expert knowl- edge of linguistics or speech technology, individuals or organizations seeking to create speech-driven applications in low- resource languages can build lexicons en- abling the recognition of small vocabular- ies (up to 100 terms, roughly) in the target language using an existing recognition en- gine designed for a high-resource source language (e.g. English). To build such lex- icons, we employ an existing method for cross-language phoneme-mapping. The application also offers a built-in audio recorder that facilitates data collection, a significantly faster implementation of the phoneme-mapp...