BnTTS: Few-Shot Speaker Adaptation in Low-Resource Setting
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929705923379200 |
|---|---|
| author | Basher, Mohammad Jahid Ibna Kowsher, Md Islam, Md Saiful Nandi, Rabindra Nath Prottasha, Nusrat Jahan Menon, Mehadi Hasan Muntasir, Tareq Al Chowdhury, Shammur Absar Alam, Firoj Yousefi, Niloofar Garibay, Ozlem Ozmen |
| author_facet | Basher, Mohammad Jahid Ibna Kowsher, Md Islam, Md Saiful Nandi, Rabindra Nath Prottasha, Nusrat Jahan Menon, Mehadi Hasan Muntasir, Tareq Al Chowdhury, Shammur Absar Alam, Firoj Yousefi, Niloofar Garibay, Ozlem Ozmen |
| contents | This paper introduces BnTTS (Bangla Text-To-Speech), the first framework for Bangla speaker adaptation-based TTS, designed to bridge the gap in Bangla speech synthesis using minimal training data. Building upon the XTTS architecture, our approach integrates Bangla into a multilingual TTS pipeline, with modifications to account for the phonetic and linguistic characteristics of the language. We pre-train BnTTS on 3.85k hours of Bangla speech dataset with corresponding text labels and evaluate performance in both zero-shot and few-shot settings on our proposed test dataset. Empirical evaluations in few-shot settings show that BnTTS significantly improves the naturalness, intelligibility, and speaker fidelity of synthesized Bangla speech. Compared to state-of-the-art Bangla TTS systems, BnTTS exhibits superior performance in Subjective Mean Opinion Score (SMOS), Naturalness, and Clarity metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_05729 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | BnTTS: Few-Shot Speaker Adaptation in Low-Resource Setting Basher, Mohammad Jahid Ibna Kowsher, Md Islam, Md Saiful Nandi, Rabindra Nath Prottasha, Nusrat Jahan Menon, Mehadi Hasan Muntasir, Tareq Al Chowdhury, Shammur Absar Alam, Firoj Yousefi, Niloofar Garibay, Ozlem Ozmen Computation and Language This paper introduces BnTTS (Bangla Text-To-Speech), the first framework for Bangla speaker adaptation-based TTS, designed to bridge the gap in Bangla speech synthesis using minimal training data. Building upon the XTTS architecture, our approach integrates Bangla into a multilingual TTS pipeline, with modifications to account for the phonetic and linguistic characteristics of the language. We pre-train BnTTS on 3.85k hours of Bangla speech dataset with corresponding text labels and evaluate performance in both zero-shot and few-shot settings on our proposed test dataset. Empirical evaluations in few-shot settings show that BnTTS significantly improves the naturalness, intelligibility, and speaker fidelity of synthesized Bangla speech. Compared to state-of-the-art Bangla TTS systems, BnTTS exhibits superior performance in Subjective Mean Opinion Score (SMOS), Naturalness, and Clarity metrics. |
| title | BnTTS: Few-Shot Speaker Adaptation in Low-Resource Setting |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2502.05729 |