Machine Learning for Audio/Speech Processing
Our research focuses on front-end audio processing techniques for speech processing, such as voice activity detection (VAD), noise suppression, direction of arrival (DoA) estimation, and classification.
![[C93] SODA: A Unified Framework for Join Estimation of Speaker Orientation and Direction of Arrival](/publication/c93_icassp_2026/featured_hued2cad710a692f8b208f3c97dce35ddf_661964_150x0_resize_q75_h2_lanczos_3.webp)
![[C91] No Encore: Unlearning as Opt-Out in Music Generation](/publication/c91_neurips_workshop_2025/featured_hu69c83db33d1b8ec3372bb284bbe7c12f_357141_150x0_resize_q75_h2_lanczos_3.webp)
![[C75] Do Not Mimic My Voice : Speaker Identity Unlearning for Zero-Shot Text-to-Speech](/publication/c75_icml_2025/featured_hu56036b6e0663b2958a1c9637bf664f91_423941_150x0_resize_q75_h2_lanczos_3.webp)
![[C40] Regression to Classification: Waveform Encoding for Neural Field-Based Audio Signal Representation](/publication/c40__icassp_2023/featured_hufe5cad49674a353f4143418c58fd15f6_72138_150x0_resize_q75_h2_lanczos_3.webp)
![[C34] NAS-VAD: Neural Architecture Search for Voice Activity Detection](/publication/c34_interspeech-2022/Featured_huc7209cf48e63204c8ba514035d8f3716_267845_150x0_resize_q75_h2_lanczos.webp)
![[C33] ADA-VAD: Unpaired Adversarial Domain Adaptation for Noise-Robust Voice Activity Detection](/publication/c33_icassp_2022/Featured_hu6aef2193e6bece41bd2a0e680fa624e7_267845_150x0_resize_q75_h2_lanczos.webp)
![[J16] ORVAE: One-Class Residual Variational Autoencoder for Voice Activity Detection in Noisy Environment](/publication/j16_npl_2022/featured_hu6aef2193e6bece41bd2a0e680fa624e7_127456_150x0_resize_q75_h2_lanczos.webp)
![[C27] Application of Adversarial Domain Adaptation to Voice Activity Detection](/publication/c27_intellisys_2021/Featured_hu6aef2193e6bece41bd2a0e680fa624e7_85515_150x0_resize_q75_h2_lanczos.webp){kind=small}
![[C17] Precision Scaling of Neural Networks for Efficient Voice Activity Detection](/publication/c17_icassp_2018_vad/featured_hu32d1899399ed0dbf38c837842d4c5814_71668_150x0_resize_q75_h2_lanczos.webp)
![[C14] Precision Scaling of Neural Networks for Efficient Audio Processing](/publication/c14_nips_2017_speech/featured_hu32d1899399ed0dbf38c837842d4c5814_51994_150x0_resize_q75_h2_lanczos.webp)