Speech Language Models (SLMs) have recently emerged as a unified paradigm for addressing a wide range of speech-related tasks, including text-to-speech (TTS), speech enhancement (SE), and automatic speech recognition (ASR). However, the generalization capability of large-scale pre-trained SLMs remains underexplored. In this work, we adapt a 1.7B parameter TTS pretrained SLM for singing voice synthesis (SVS), using only a 135-hour synthetic singing corpus, ACE-Opencpop. Building upon the ESPNet-SpeechLM, our recipe involves the following procedure: (1) tokenization of music score conditions and singing waveforms, (2) multi-stream language model token prediction, (3) conditional flow matching-based mel-spectrogram generation. (4) a mel-to-wave vocoder. Experimental results demonstrate that our adapted SLM generalizes well to SVS and achieves performance comparable to leading discrete token-based SVS models.
@article{arxiv.2512.14657,
title = {Adapting Speech Language Model to Singing Voice Synthesis},
author = {Yiwen Zhao and Jiatong Shi and Jinchuan Tian and Yuxun Tang and Jiarui Hai and Jionghao Han and Shinji Watanabe},
journal= {arXiv preprint arXiv:2512.14657},
year = {2025}
}