Denoising Diffusion Conditional Expectation Model
for Efficient and Robust Target Speech Extraction

Target Speech Extraction (TSE) is a crucial task in speech processing that focuses on isolating the clean speech of a specific speaker from complex mixtures. While discriminative methods are commonly used for TSE, they can introduce distortion in terms of speech perception quality. On the other hand, generative approaches, particularly diffusion-based methods, can enhance speech quality perceptually but suffer from slower inference speed. We propose an efficient generative approach named Denoising Diffusion Conditional Expectation Model (DDCEM) for TSE. It can handle multi- and single-speaker scenarios in both noisy and clean conditions. Additionally, we introduce Regenerate-DDCEM (R-DDCEM) that can regenerate and optimize speech quality based on pre-processed speech from a discriminative model. Our method outperforms conventional methods in terms of both intrusive and non-intrusive metrics and demonstrates notable strengths in inference efficiency and robustness to unseen tasks.

Mixture: The input multi-speaker mixture with background noise.
Ground Truth: The clean speech of the target speaker.
DDCEM: Extraction results of our proposed DDCEM method.
DPCCN: Extraction results of discriminative DPCCN method.
DiffSep+SV: Extraction results of diffusion-based DiffSep+SV method.

Mixture: The input multi-speaker mixture without background noise.
Ground Truth: The clean speech of the target speaker.
DDCEM: Extraction results of our proposed DDCEM method.
DPCCN: Extraction results of discriminative DPCCN method.
DiffSep+SV: Extraction results of diffusion-based DiffSep+SV method.

Mixture: The input noisy single-speaker speech.
Ground Truth: The clean speech of the target speaker.
DDCEM: Enhancement results of our proposed DDCEM method.
DCCRN: Enhancement results of discriminative DPCCN method.
SGMSE+: Enhancement results of diffusion based SGMSE+ method.