MASRI-HEADSET: A Maltese Corpus for Speech Recognition

被引：0

作者：

Mena, Carlos ^{[1
]}

Gatt, Albert ^{[1
]}

DeMarco, Andrea ^{[1
]}

Borg, Claudia ^{[1
]}

van der Plas, Lonneke ^{[1
]}

Muscat, Amanda ^{[1
]}

Padovani, Ian ^{[1
]}

机构：

[1] Univ Malta, Msida, Malta

来源：

PROCEEDINGS OF THE 12TH INTERNATIONAL CONFERENCE ON LANGUAGE RESOURCES AND EVALUATION (LREC 2020) | 2020年

关键词：

Maltese; Speech Corpora; Automatic Speech Recognition;

D O I：

暂无

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

Maltese, the national language of Malta, is spoken by approximately 500,000 people. Speech processing for Maltese is still in its early stages of development. In this paper, we present the first spoken Maltese corpus designed purposely for Automatic Speech Recognition (ASR). The MASRI-HEADSET corpus was developed by the MASRI project at the University of Malta. It consists of 8 hours of speech paired with text, recorded by using short text snippets in a laboratory environment. The speakers were recruited from different geographical locations all over the Maltese islands, and were roughly evenly distributed by gender. This paper also presents some initial results achieved in baseline experiments for Maltese ASR using Sphinx and Kaldi. The MASRI-HEADSET Corpus is publicly available for research/academic purposes.

引用

页码：6381 / 6388

页数：8

共 40 条

[11]

Ceplo S, 2018, THESIS

[12]

Comrie B., 2009, INTRO MALTESE LINGUI, P3, DOI [10.1075/slcs.113.04com, DOI 10.1075/SLCS.113.04COM]

[13]

Draxler C., 2004, SPEECH RECORDER VERS

[14]

Fabri R., 1993, KONGRUENZ GRAMMATIK

[15]

Gatt A., 2018, The Languages of Malta, P143, DOI DOI 10.5281/ZENODO.1181793

[16]

Gatt Albert, 2013, P INT C CORPUS LING, P96

[17]

Grech S., 2015, THESIS

[18]

Hannun A., 2014, CoRR

[19]

Hoberman R.D., 2007, MORPHOLOGIES ASIA AF, V1, P257, DOI [10.5325/j.ctv1bxh537.18, DOI 10.5325/J.CTV1BXH537.18]

[20]

Huggins-Daines D, 2006, INT CONF ACOUST SPEE, P185

← 1 2 3 4 →