Cross-lingual speech recognition between languages from the same language family

Proceedings of the Romanian Academy Series A-Mathematics Physics Technical Sciences Information Science

Andrej ZGANK¹ 

关键词: automatic speech recognition;    cross-lingual phoneme mapping;    expert-driven approach;    similar languages.;   

DOI  :  

学科分类：计算机科学（综合）

来源: Editura Academiei Romane

【 摘 要 】

This paper presents a detailed analysis of how languages, members of different languagesubfamilies, influence word accuracy in the case of cross-lingual speech recognition. Cross-lingualspeech recognition presents an efficient method for developing an automatic speech recognitionsystem for ICT applications and services in the case of under-resourced languages, where the majorityof Eastern and Central European languages can be classified. In the Romance language familyexperiments, Spanish was selected as the source and Romanian as the target language. Theexperiments were carried out using the Spanish SpeechDat(II) speech database and a dedicatedRomanian speech database. The Spanish source automatic speech recognition system was trainedusing the MASPER system. A cross-lingual phoneme mapping approach based on expert knowledgewas proposed for the Romanian-Spanish language pair. The Romanian cross-lingual speechrecognition system achieved word accuracy of 80.48 % .

【 授权许可】

Unknown