Diffusion Posterior Sampling for Informed Single-Channel Dereverberation
This work addresses speech quality enhancement in reverberant environments, such as teleconferencing or hearing aids, by providing an incremental improvement over existing informed dereverberation techniques.
The authors tackled single-channel speech dereverberation by proposing a diffusion-based method that uses knowledge of the room impulse response to generate clean speech, showing improved robustness to noise, especially non-stationary types, and superiority over other methods for large reverberation times.
We present in this paper an informed single-channel dereverberation method based on conditional generation with diffusion models. With knowledge of the room impulse response, the anechoic utterance is generated via reverse diffusion using a measurement consistency criterion coupled with a neural network that represents the clean speech prior. The proposed approach is largely more robust to measurement noise compared to a state-of-the-art informed single-channel dereverberation method, especially for non-stationary noise. Furthermore, we compare to other blind dereverberation methods using diffusion models and show superiority of the proposed approach for large reverberation times. We motivate the presented algorithm by introducing an extension for blind dereverberation allowing joint estimation of the room impulse response and anechoic speech. Audio samples and code can be found online (https://uhh.de/inf-sp-derev-dps).