AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

EA-VAE: Learning to Reconstruct Dysarthric Speech via Variational Autoencoder with Encoding Alignment

Daipeng Zhang, Wenhuan Lu, Xianghu Yue, Hongcheng Zhang, Jianguo Wei

AAAI 2026 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Regularizing VAE latent information · 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 Distribution Alignment Module (KL between dysarthric and normal priors) (ablation)EA-VAE · UASpeech B2 averageWER (%): 94.46 -> 62.19Improved
add Distribution Alignment Module (ablation)EA-VAE · UASpeech B2 averageMOS: 1.66 -> 4.48Improved
add Embedding Alignment Module (ablation)EA-VAE · UASpeech B2 averageWER (%): 71.75 -> 62.19Improved
add Embedding Alignment Module (ablation)EA-VAE · UASpeech B2 averageMOS: 3.56 -> 4.48Improved
add Duration Alignment Predictor (ablation)EA-VAE · UASpeech B2 averageWER (%): 63.43 -> 62.19Improved
add Duration Alignment Predictor (ablation)EA-VAE · UASpeech B2 averageMOS: 4.13 -> 4.48Improved
time-stretch dysarthric audio so Td >= Tn (vs Td < Tn) (ablation)EA-VAE · UASpeech B2 averageWER (%): 83.60 -> 62.19Improved
time-stretch dysarthric audio so Td >= Tn (vs Td < Tn) (ablation)EA-VAE · UASpeech B2 averageMOS: 1.23 -> 4.48Improved
use EA-VAE instead of Unit-DSRdysarthric speech reconstruction · UASpeech B2 averageWER (%): 65.08 -> 62.19Improved
use EA-VAE instead of ASR-TTS on speaker F02 MOSdysarthric speech reconstruction · UASpeech F02 (Low)MOS: 4.02 vs Unit-DSR 4.42Mixed

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