BnTTS: Few-Shot Speaker Adaptation in Low-Resource Setting

2025/02/09 by Mohammad Jahid Ibna Basher, Md. Kowsher, Basher, Mohammad Jahid Ibna +19
Computer Science · #Advanced Data Compression Techniques #Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · pdf · doi:10.48550/arxiv.2502.05729

openalex publication_date 2025/02/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

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.

Related

• A2TTS: TTS for Low Resource Indian Languages
• End to End Bangla Speech Synthesis
• Byakto Speech: Real-time long speech synthesis with convolutional neural network: Transfer learning from English to Bangla
• Bangla-Wave: Improving Bangla Automatic Speech Recognition Utilizing N-gram Language Models
• Kinship in Speech: Leveraging Linguistic Relatedness for Zero-Shot TTS in Indian Languages