Saved in:
Bibliographic Details
Main Authors: Mei, Jiahao, Xu, Xuenan, Xie, Zeyu, Zheng, Zihao, Tao, Ye, Ding, Yue, Wu, Mengyue
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.05875
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917329211752448
author Mei, Jiahao
Xu, Xuenan
Xie, Zeyu
Zheng, Zihao
Tao, Ye
Ding, Yue
Wu, Mengyue
author_facet Mei, Jiahao
Xu, Xuenan
Xie, Zeyu
Zheng, Zihao
Tao, Ye
Ding, Yue
Wu, Mengyue
contents Recent advances in text-to-music models have enabled coherent music generation from text prompts, yet fine-grained emotional control remains unresolved. We introduce LARA-Gen, a framework for continuous emotion control that aligns the internal hidden states with an external music understanding model through Latent Affective Representation Alignment (LARA), enabling effective training. In addition, we design an emotion control module based on a continuous valence-arousal space, disentangling emotional attributes from textual content and bypassing the bottlenecks of text-based prompting. Furthermore, we establish a benchmark with a curated test set and a robust Emotion Predictor, facilitating objective evaluation of emotional controllability in music generation. Extensive experiments demonstrate that LARA-Gen achieves continuous, fine-grained control of emotion and significantly outperforms baselines in both emotion adherence and music quality. Generated samples are available at https://anonymous2232330.github.io/laragen-web/.
format Preprint
id arxiv_https___arxiv_org_abs_2510_05875
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LARA-Gen: Enabling Continuous Emotion Control for Music Generation Models via Latent Affective Representation Alignment
Mei, Jiahao
Xu, Xuenan
Xie, Zeyu
Zheng, Zihao
Tao, Ye
Ding, Yue
Wu, Mengyue
Sound
Recent advances in text-to-music models have enabled coherent music generation from text prompts, yet fine-grained emotional control remains unresolved. We introduce LARA-Gen, a framework for continuous emotion control that aligns the internal hidden states with an external music understanding model through Latent Affective Representation Alignment (LARA), enabling effective training. In addition, we design an emotion control module based on a continuous valence-arousal space, disentangling emotional attributes from textual content and bypassing the bottlenecks of text-based prompting. Furthermore, we establish a benchmark with a curated test set and a robust Emotion Predictor, facilitating objective evaluation of emotional controllability in music generation. Extensive experiments demonstrate that LARA-Gen achieves continuous, fine-grained control of emotion and significantly outperforms baselines in both emotion adherence and music quality. Generated samples are available at https://anonymous2232330.github.io/laragen-web/.
title LARA-Gen: Enabling Continuous Emotion Control for Music Generation Models via Latent Affective Representation Alignment
topic Sound
url https://arxiv.org/abs/2510.05875