AudioX-Turbo: Unified Framework for Efficient Anything-to-Audio Generation

Arxiv CS.SD (Sound/Audio ML)

AudioX-Turbo presents a unified multimodal framework for efficient audio/music generation from text, images, and reference audio. It addresses key challenges in modeling, large-scale training data, and inference cost.

See 6 more sources

PianoKontext: Expressive Performance Rendering from Deadpan Context3 days agoDmitrii Gavrilev

A Sensitivity Analysis of Multi-Event Audio Grounding in Audio LLMs3 days agoTaehan Lee, Jaehan Jung, Hyukjun Lee

Snapping Matters: Context-Aware Onset Refinement for Automatic Music Transcription3 days agoAbhirup Saha, Hans-Ulrich Berendes, Meinard M\"uller, Ben Maman

Can LLMs understand LilyPond? A benchmark for symbolic music generation and understanding5 days agoMatteo Spanio, Mohammad Torabi, Andrea Poltronieri, Antonio Rod\`a

Real-Time Language Model Jamming: A Case Study for Live Music Accompaniment Generation3 days agoBowen Zheng, Andrew H. Yang, Jiaqi Ruan, Jia He, Xinyue Li, Yuan-Hsin Chen, Ziyu Wang, Xiaosong Ma

AudioX-Turbo: A Unified Framework for Efficient Anything-to-Audio Generation2 days agoZeyue Tian, Lei Ke, Zhaoyang Liu, Ruibin Yuan, Liumeng Xue, Yujiu Yang, Weijia Chen, Xu Tan, Qifeng Chen, Wei Xue, Yike Guo

Arxiv CS.SD (Sound/Audio ML)·Joonyong ParkJungwoo KimJunyoung KohYuki Saito

2 days ago