AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

MF-Speech: Achieving Fine-Grained and Compositional Control in Speech Generation via Factor Disentanglement

Xinyue Yu, Youqing Fang, Pingyu Wu, Guoyang Ye, Wenbo Zhou, Weiming Zhang, Song Xiao

AAAI 2026 · Image generation › Autoregressive and latent generation · approach: Neural text-to-speech synthesis · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add dynamic fusion (DyGate) (ablation)MF-SpeechGenerator · ESD multi-factor compositional generationSECS: 0.5551 -> 0.5685Improved
add dynamic fusion (DyGate) (ablation)MF-SpeechGenerator · ESD multi-factor compositional generationWER: 5.17% -> 4.76%Improved
add dynamic fusion (DyGate) (ablation)MF-SpeechGenerator · ESD multi-factor compositional generationUTMOS, Log RMSE, CorrNo clear effect
add HSAN (ablation)MF-SpeechGenerator · ESD multi-factor compositional generationSECS: 0.1576 -> 0.5685Improved
add HSAN (ablation)MF-SpeechGenerator · ESD multi-factor compositional generationCorr: 0.64 -> 0.68Improved
add HSAN (ablation)MF-SpeechGenerator · ESD multi-factor compositional generationLog RMSE: 0.38 -> 0.34Improved
add HSAN (ablation)MF-SpeechGenerator · ESD multi-factor compositional generationWERHurt
add mutual information constraints (ablation)MF-SpeechEncoder · ESD t-SNE of timbre and emotioncluster separabilityImproved
add contrastive learning (ablation)MF-SpeechEncoder · ESD t-SNE of timbre and emotioncluster separabilityImproved
add prosody priors (ablation)MF-SpeechEncoder · ESD t-SNE of timbre and emotioncluster separabilityImproved
use MF-SpeechEncoder instead of FACodec encoderdisentanglement · ESD factor disentanglementAcc_e: 0.2343 -> 0.9296Improved
use MF-Speech instead of DDDM-VCspeech generation · ESD multi-factor compositional generationWER: 11.67% -> 4.67%Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy