Buchumschlag

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Hu, Jingbin, Chen, Huakang, Ma, Linhan, Guo, Dake, Zhan, Qirui, Li, Wenhao, Zhang, Haoyu, Xia, Kangxiang, Zhang, Ziyu, Tian, Wenjie, Wang, Chengyou, Liang, Jinrui, Guo, Shuhan, Yang, Zihang, Wu, Bengu, Zhang, Binbin, Zhu, Pengcheng, Xie, Pengyuan, Xie, Chuan, Zhang, Qiang, Liu, Jie, Xie, Lei
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Audio and Speech Processing
Online-Zugang:https://arxiv.org/abs/2601.10629
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Inhaltsangabe:
  • Despite rapid progress in text-to-speech (TTS), open-source systems still lack truly instruction-following, fine-grained control over core speech attributes (e.g., pitch, speaking rate, age, emotion, and style). We present VoiceSculptor, an open-source unified system that bridges this gap by integrating instruction-based voice design and high-fidelity voice cloning in a single framework. It generates controllable speaker timbre directly from natural-language descriptions, supports iterative refinement via Retrieval-Augmented Generation (RAG), and provides attribute-level edits across multiple dimensions. The designed voice is then rendered into a prompt waveform and fed into a cloning model to enable high-fidelity timbre transfer for downstream speech synthesis. VoiceSculptor achieves open-source state-of-the-art (SOTA) on InstructTTSEval-Zh, and is fully open-sourced, including code and pretrained models, to advance reproducible instruction-controlled TTS research.

Ähnliche Einträge