7 papers
SD2AIL: Adversarial Imitation Learning from Synthetic Demonstrations via Diffusion Models
Pengcheng Li, Qiang Fang, Tong Zhao +2
Adversarial Imitation Learning (AIL) is a dominant framework in imitation learning that infers rewards from expert demonstrations to guide policy optimization. Although providing m…
EMO-RL: Emotion-Rule-Based Reinforcement Learning Enhanced Audio-Language Model for Generalized Speech Emotion Recognition
Pengcheng Li, Botao Zhao, Zuheng Kang +4
Although Large Audio-Language Models (LALMs) have exhibited outstanding performance in auditory understanding, their performance in affective computing scenarios, particularly in e…
IDEAW: Robust Neural Audio Watermarking with Invertible Dual-Embedding
Pengcheng Li, Xulong Zhang, Jing Xiao +1
The audio watermarking technique embeds messages into audio and accurately extracts messages from the watermarked audio. Traditional methods develop algorithms based on expert expe…
MAIN-VC: Lightweight Speech Representation Disentanglement for One-shot Voice Conversion
Pengcheng Li, Jianzong Wang, Xulong Zhang +3
One-shot voice conversion aims to change the timbre of any source speech to match that of the unseen target speaker with only one speech sample. Existing methods face difficulties…
CONTUNER: Singing Voice Beautifying with Pitch and Expressiveness Condition
Jianzong Wang, Pengcheng Li, Xulong Zhang +2
Singing voice beautifying is a novel task that has application value in people's daily life, aiming to correct the pitch of the singing voice and improve the expressiveness without…
Medical Speech Symptoms Classification via Disentangled Representation
Jianzong Wang, Pengcheng Li, Xulong Zhang +2
Intent is defined for understanding spoken language in existing works. Both textual features and acoustic features involved in medical speech contain intent, which is important for…
DQR-TTS: Semi-supervised Text-to-speech Synthesis with Dynamic Quantized Representation
Jianzong Wang, Pengcheng Li, Xulong Zhang +2
Most existing neural-based text-to-speech methods rely on extensive datasets and face challenges under low-resource condition. In this paper, we introduce a novel semi-supervised t…