Virtual microphones enhance blind speech separation with diffusion models
VM-ARRAYDPS: Virtual Microphone Augmented Diffusion Posterior Sampling for Unsupervised Blind Speech Separation
Machine LearningMultimediaSound
Summary
Separating multiple people's voices from a noisy recording without knowing anything about the original sounds is a hard problem. The authors improve a recent method that uses smart AI models and multiple microphones by creating extra 'virtual' microphones with clearer sound. These virtual microphones give more clues to help separate the voices better. Their experiments show this new approach works better when separating two or three speakers from recordings.
What this means in practice
- •For audio engineers: Improve multi-speaker speech separation in recordings by augmenting physical microphones with virtual ones for clearer source recovery.
- •For telecommunications developers: Enhance speech clarity in multi-microphone communication devices by integrating virtual microphone constraints to better separate overlapping speakers.
Authors
Jingqi Sun, Haozhan Tang, Shulin He, Zhong-Qiu Wang
Abstract
Blind Source Separation(BSS) is a fundamental problem in signal processing, aiming to separate multiple source signals from their mixtures without prior knowledge of the sources or the mixing process. Traditional approaches, such as Independent Vector Analysis (IVA) exploits statistical independence of sources. Recently, diffusion-based approaches have emerged as a promising alternative by leveraging powerful generative priors. Among them, ArrayDPS formulates BSS problem as a posterior sampling problem, and utilizes a pretrained speech diffusion model to guide the recovery of clean source signals. A key factor behind its separation capability is the multi-channel consistency (MC) objective, which enforces the estimated source signals to reconstruct the observed microphone mixtures through the estimated acoustic transfer functions. However, the number of microphones in the array is often limited, which constrains the performance of ArrayDPS. To address this issue, we propose VM-ArrayDPS, a novel method that augments the microphone array with virtual microphones with higher-SNR, these microphones can offer extra MC constraints to enhance the separation performance. Experimental results demonstrate that VM-ArrayDPS significantly outperforms ArrayDPS on both 2-speaker and 3-speaker datasets, showcasing the effectiveness of virtual microphone augmentation in improving BSS performance. We also did ablation studies to show the influence of the number of virtual microphones and weight of the MC objective brought by virtual microphones.