First Automatic Fongbe Continuous Speech Recognition System: Development of Acoustic Models and Language Models

Abstract : This paper reports our efforts toward an ASR system for a new under-resourced language (Fongbe). The aim of this work is to build acoustic models and language models for continuous speech decoding in Fongbe. The problem encountered with Fongbe (an African language spoken especially in Benin, Togo, and Nigeria) is that it does not have any language resources for an ASR system. As part of this work, we have first collected Fongbe text and speech corpora that are described in the following sections. Acoustic modeling has been worked out at a graphemic level and language modeling has provided two language models for performance comparison purposes. We also performed a vowel simplification by removing tones diacritics in order to investigate their impact on the language models.
Document type :
Conference papers
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01436788
Contributor : Laurent Besacier <>
Submitted on : Saturday, January 21, 2017 - 7:16:48 AM
Last modification on : Thursday, April 4, 2019 - 10:18:05 AM
Document(s) archivé(s) le : Saturday, April 22, 2017 - 12:24:55 PM

File

153.pdf
Publisher files allowed on an open archive

Identifiers

Collections

Citation

Fréjus Laleye, Laurent Besacier, Eugène Ezin, Cina Motamed. First Automatic Fongbe Continuous Speech Recognition System: Development of Acoustic Models and Language Models. 2016 Federated Conference on Computer Science and Information Systems (FedCSIS), Sep 2016, Gdansk, Poland. pp.477 - 482, ⟨10.15439/2016F153⟩. ⟨hal-01436788⟩

Share

Metrics

Record views

157

Files downloads

212