17 citations · 80 across the 19 of their papers we have counts for
14 papers · 1 filter
SymPAC: Scalable Symbolic Music Generation With Prompts And Constraints
Haonan Chen, Jordan B. L. Smith, Janne Spijkervet +5
Progress in the task of symbolic music generation may be lagging behind other tasks like audio and text generation, in part because of the scarcity of symbolic training data. In th…
ByteComposer: a Human-like Melody Composition Method based on Language Model Agent
Xia Liang, Xingjian Du, Jiaju Lin +3
Large Language Models (LLM) have shown encouraging progress in multimodal understanding and generation tasks. However, how to design a human-aligned and interpretable melody compos…
Joint Music and Language Attention Models for Zero-shot Music Tagging
Xingjian Du, Zhesong Yu, Jiaju Lin +2
Music tagging is a task to predict the tags of music recordings. However, previous music tagging research primarily focuses on close-set music tagging tasks which can not be genera…
Universal Source Separation with Weakly Labelled Data
Qiuqiang Kong, Ke Chen, Haohe Liu +4
Universal source separation (USS) is a fundamental research task for computational auditory scene analysis, which aims to separate mono recordings into individual source tracks. Th…
ByteCover3: Accurate Cover Song Identification on Short Queries
Xingjian Du, Zijie Wang, Xia Liang +3
Deep learning based methods have become a paradigm for cover song identification (CSI) in recent years, where the ByteCover systems have achieved state-of-the-art results on all th…
HTS-AT: A Hierarchical Token-Semantic Audio Transformer for Sound Classification and Detection
Ke Chen, Xingjian Du, Bilei Zhu +3
Audio classification is an important task of mapping audio samples into their corresponding labels. Recently, the transformer model with self-attention mechanisms has been adopted…
Zero-shot Audio Source Separation through Query-based Learning from Weakly-labeled Data
Ke Chen, Xingjian Du, Bilei Zhu +3
Deep learning techniques for separating audio into different sound sources face several challenges. Standard architectures require training separate models for different types of a…
Attention-based cross-modal fusion for audio-visual voice activity detection in musical video streams
Yuanbo Hou, Zhesong Yu, Xia Liang +4
Many previous audio-visual voice-related works focus on speech, ignoring the singing voice in the growing number of musical video streams on the Internet. For processing diverse mu…
Speech enhancement with weakly labelled data from AudioSet
Qiuqiang Kong, Haohe Liu, Xingjian Du +3
Speech enhancement is a task to improve the intelligibility and perceptual quality of degraded speech signal. Recently, neural networks based methods have been applied to speech en…
CatNet: music source separation system with mix-audio augmentation
Xuchen Song, Qiuqiang Kong, Xingjian Du +1
Music source separation (MSS) is the task of separating a music piece into individual sources, such as vocals and accompaniment. Recently, neural network based methods have been ap…
Contrastive Unsupervised Learning for Audio Fingerprinting
Zhesong Yu, Xingjian Du, Bilei Zhu +1
The rise of video-sharing platforms has attracted more and more people to shoot videos and upload them to the Internet. These videos mostly contain a carefully-edited background au…
ByteCover: Cover Song Identification via Multi-Loss Training
Xingjian Du, Zhesong Yu, Bilei Zhu +2
We present in this paper ByteCover, which is a new feature learning method for cover song identification (CSI). ByteCover is built based on the classical ResNet model, and two majo…
End-to-End Model for Speech Enhancement by Consistent Spectrogram Masking
Xingjian Du, Mengyao Zhu, Xuan Shi +3
Recently, phase processing is attracting increasinginterest in speech enhancement community. Some researchersintegrate phase estimations module into speech enhancementmodels by usi…
End-to-End Residual CNN with L-GM Loss Speaker Verification System
Xuan Shi, Xingjian Du, Mengyao Zhu
We propose an end-to-end speaker verification system based on the neural network and trained by a loss function with less computational complexity. The end-to-end speaker verificat…