A multilingual training strategy for low resource Text to Speech
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913490397036544 |
|---|---|
| author | Amalas, Asma Ghogho, Mounir Chetouani, Mohamed Thami, Rachid Oulad Haj |
| author_facet | Amalas, Asma Ghogho, Mounir Chetouani, Mohamed Thami, Rachid Oulad Haj |
| contents | Recent speech technologies have led to produce high quality synthesised speech due to recent advances in neural Text to Speech (TTS). However, such TTS models depend on extensive amounts of data that can be costly to produce and is hardly scalable to all existing languages, especially that seldom attention is given to low resource languages. With techniques such as knowledge transfer, the burden of creating datasets can be alleviated. In this paper, we therefore investigate two aspects; firstly, whether data from social media can be used for a small TTS dataset construction, and secondly whether cross lingual transfer learning (TL) for a low resource language can work with this type of data. In this aspect, we specifically assess to what extent multilingual modeling can be leveraged as an alternative to training on monolingual corporas. To do so, we explore how data from foreign languages may be selected and pooled to train a TTS model for a target low resource language. Our findings show that multilingual pre-training is better than monolingual pre-training at increasing the intelligibility and naturalness of the generated speech. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_01217 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A multilingual training strategy for low resource Text to Speech Amalas, Asma Ghogho, Mounir Chetouani, Mohamed Thami, Rachid Oulad Haj Computation and Language Machine Learning Sound Audio and Speech Processing Recent speech technologies have led to produce high quality synthesised speech due to recent advances in neural Text to Speech (TTS). However, such TTS models depend on extensive amounts of data that can be costly to produce and is hardly scalable to all existing languages, especially that seldom attention is given to low resource languages. With techniques such as knowledge transfer, the burden of creating datasets can be alleviated. In this paper, we therefore investigate two aspects; firstly, whether data from social media can be used for a small TTS dataset construction, and secondly whether cross lingual transfer learning (TL) for a low resource language can work with this type of data. In this aspect, we specifically assess to what extent multilingual modeling can be leveraged as an alternative to training on monolingual corporas. To do so, we explore how data from foreign languages may be selected and pooled to train a TTS model for a target low resource language. Our findings show that multilingual pre-training is better than monolingual pre-training at increasing the intelligibility and naturalness of the generated speech. |
| title | A multilingual training strategy for low resource Text to Speech |
| topic | Computation and Language Machine Learning Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2409.01217 |