Using Resources from a Closely-related Language to Develop ASR for a Very Under-resourced Language: A Case Study for Iban
This paper presents our strategies for developing an automatic speech recognition system for Iban, an under-resourced language. We faced several challenges such as no pronunciation dictionary and lack of training material for building acoustic models. To overcome these problems, we proposed approach...
| Main Authors: | , , , |
|---|---|
| Format: | Proceeding |
| Language: | English |
| Published: |
2015
|
| Subjects: | |
| Online Access: | http://ir.unimas.my/id/eprint/8883/ http://ir.unimas.my/id/eprint/8883/1/IS2015_samsonjuan_camera-ready.pdf |
| _version_ | 1848836463648571392 |
|---|---|
| author | Juan, Sarah Samson Besacier, Laurent Lecouteux, Benjamin Dyab, Mohamed |
| author_facet | Juan, Sarah Samson Besacier, Laurent Lecouteux, Benjamin Dyab, Mohamed |
| author_sort | Juan, Sarah Samson |
| building | UNIMAS Institutional Repository |
| collection | Online Access |
| description | This paper presents our strategies for developing an automatic speech recognition system for Iban, an under-resourced language. We faced several challenges such as no pronunciation dictionary and lack of training material for building acoustic models. To overcome these problems, we proposed approaches which exploit resources from a closely-related language (Malay). We developed a semi-supervised method for building the pronunciation dictionary and applied cross-lingual strategies for improving acoustic models trained with very limited training data. Both approaches displayed very encouraging results, which show that data from a closely-related language, if available, can be exploited to build ASR for a new language. In the final part of the paper, we present a zero-shot ASR using Malay resources that can be used as an alternative method for transcribing Iban speech. |
| first_indexed | 2025-11-15T06:24:10Z |
| format | Proceeding |
| id | unimas-8883 |
| institution | Universiti Malaysia Sarawak |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T06:24:10Z |
| publishDate | 2015 |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | unimas-88832015-10-16T01:23:21Z http://ir.unimas.my/id/eprint/8883/ Using Resources from a Closely-related Language to Develop ASR for a Very Under-resourced Language: A Case Study for Iban Juan, Sarah Samson Besacier, Laurent Lecouteux, Benjamin Dyab, Mohamed QA75 Electronic computers. Computer science This paper presents our strategies for developing an automatic speech recognition system for Iban, an under-resourced language. We faced several challenges such as no pronunciation dictionary and lack of training material for building acoustic models. To overcome these problems, we proposed approaches which exploit resources from a closely-related language (Malay). We developed a semi-supervised method for building the pronunciation dictionary and applied cross-lingual strategies for improving acoustic models trained with very limited training data. Both approaches displayed very encouraging results, which show that data from a closely-related language, if available, can be exploited to build ASR for a new language. In the final part of the paper, we present a zero-shot ASR using Malay resources that can be used as an alternative method for transcribing Iban speech. 2015-09 Proceeding PeerReviewed text en http://ir.unimas.my/id/eprint/8883/1/IS2015_samsonjuan_camera-ready.pdf Juan, Sarah Samson and Besacier, Laurent and Lecouteux, Benjamin and Dyab, Mohamed (2015) Using Resources from a Closely-related Language to Develop ASR for a Very Under-resourced Language: A Case Study for Iban. In: Proceedings of INTERSPEECH 2015, September 2015, Dresden, Germany. |
| spellingShingle | QA75 Electronic computers. Computer science Juan, Sarah Samson Besacier, Laurent Lecouteux, Benjamin Dyab, Mohamed Using Resources from a Closely-related Language to Develop ASR for a Very Under-resourced Language: A Case Study for Iban |
| title | Using Resources from a Closely-related Language to Develop ASR for a Very Under-resourced Language: A Case Study for Iban |
| title_full | Using Resources from a Closely-related Language to Develop ASR for a Very Under-resourced Language: A Case Study for Iban |
| title_fullStr | Using Resources from a Closely-related Language to Develop ASR for a Very Under-resourced Language: A Case Study for Iban |
| title_full_unstemmed | Using Resources from a Closely-related Language to Develop ASR for a Very Under-resourced Language: A Case Study for Iban |
| title_short | Using Resources from a Closely-related Language to Develop ASR for a Very Under-resourced Language: A Case Study for Iban |
| title_sort | using resources from a closely-related language to develop asr for a very under-resourced language: a case study for iban |
| topic | QA75 Electronic computers. Computer science |
| url | http://ir.unimas.my/id/eprint/8883/ http://ir.unimas.my/id/eprint/8883/1/IS2015_samsonjuan_camera-ready.pdf |