Research Projects

Onset-and-Offset-Aware Sound Event Detection via Differentiable Frame-to-Event Mapping

2 mins

SED-HSMM couples a neural acoustic model with a hidden semi-Markov model to deliver event-wise sound detection with precise boundaries.

Investigation of Spatial Self-Supervised Learning for Target Speaker ASR

2 mins

Guided factor models that leverage spatial self-supervision to separate target speakers for CHiME-8 DASR benchmarks.

Neural FCASA for Conversation Analysis

2 mins

Neural blind source separation and diarization for distant speech recognition that works with weak supervision.

Time-Varying Neural FCA

3 mins

This paper presents an unsupervised multichannel method that can separate moving sound sources based on an amortized variational inference (AVI) of joint separation and localization.

Neural Full-Rank Spatial Covariance Analysis

2 mins

This paper describes a neural blind source separation (BSS) method based on amortized variational inference (AVI) of a non-linear generative model of mixture signals.

Adaptive Neural Speech Enhancement with a Denoising Variational Autoencoder

2 mins

This paper presents a neural speech enhancement method that has a statistical feedback mechanism based on a denoising variational autoencoder (VAE).

Speech Enhancement Based on Robust NTF

2 mins

This paper presents a blind multichannel speech enhancement method that can deal with the time-varying layout of microphones and sound sources.