ProSing: Zero-Shot Fine-Grained Singing Technique Conversion
Xinhao Guan, Jingyue Huang, Kangrui Chen, Julian McAuley, Zhong-Qiu Wang
Abstract
We study zero-shot fine-grained singing technique conversion (STC), which applies specified techniques to selected regions of an existing singing recording while preserving lyrics, melody, timing, and singer identity. Existing singing conversion and synthesis methods offer complementary strengths in source preservation and explicit technique control, but combining both in an audio-to-audio setting remains underexplored, especially with limited technique-annotated data. We propose ProSing, a zero-shot fine-grained STC framework that adapts a pretrained flow-matching acoustic model with frame-level technique conditioning. To mitigate conditioning shortcuts during adaptation, we introduce a paired training strategy to learn generalizable technique representations. We further incorporate continuous pitch trajectories for explicit and expressive control of pitch-related techniques. Subjective and objective evaluations on unseen singers and source recordings show that ProSing achieves accurate technique control while preserving singing naturalness and singer similarity.
Create a lesson
Related papers
ZoionNET: Low-Power Animal Intrusion Detection Using an Acoustic Deep-Learning Classifier for a Cellular Remote-Sensing System
Prasanna Vijay, Adarsh Gupta, Vansh Panwar et al.
When Sounds Turn Heads: Recovering Event Timing That Audio LLMs Fail to Say
Jongbhin Woo, Jaewook Jeong, Joon Son Chung
From Attention to Semantics: Detecting and Mitigating Sound Event Hallucinations in Large Audio-Language Models
Tianyi Sun, Minghao Lv, Chunyang Yu et al.
The landscape of AI speech generation and deepfake detection
Simon King
Exploring Speaker-Aware Engagement Modeling for Children with Autism Spectrum Disorder
Kaleen Shrestha, Austin Joel Dympep, Young Suh Hong et al.
Harmonic Circuit: Chord Progression Lies in a Sparse Principal Information Flow in Symbolic Music Models
Ruiji Yu, Gus Xia