CONF ElHajal_IEEESLT_2026/IDIAP Concatenative Neural Speech Synthesis El Hajal, Karl Kumar Agrawal, Sanat Hermann, Enno Magimai-Doss, Mathew EXTERNAL https://publications.idiap.ch/attachments/papers/2026/ElHajal_IEEESLT_2026.pdf PUBLIC IEEE SLT 2026 Speech synthesis has evolved from knowledge-driven methods, such as concatenative speech synthesis, to data-driven ones, such as neural speech synthesis. Both can yield commercial-grade speech synthesis systems. In this paper, we propose a target-domain training-free multi-speaker approach that combines concatenative speech synthesis with a recently proposed neural method based on self-supervised learning representations. Through experiments on multiple voices in the CMU ARCTIC corpus, we show that this combination opens new ways to exploit prior knowledge and auxiliary data to reduce target data requirements and develop interpretable and controllable neural speech synthesis systems. Subjective and objective evaluations confirm improved naturalness compared to concatenative synthesis alone, at the cost of intelligibility.