# Block-Online Multi-Channel Speech Enhancement Using DNN-Supported Relative Transfer Function Estimates

## TL;DR

This paper proposes a block-online multi-channel speech enhancement method that uses DNN-supported relative transfer function estimates, demonstrating robustness and significant improvements in speech quality and recognition accuracy even with very short processing blocks.

## Contribution

It introduces a novel block-online approach for multi-channel speech enhancement using DNN-supported RTF estimates, suitable for dynamic environments with short utterances.

## Key findings

- Robustness of the method with short blocks of 250 ms.
- Significant improvements in speech quality metrics like PESQ.
- Enhanced automatic speech recognition performance.

## Abstract

This work addresses the problem of block-online processing for multi-channel speech enhancement. Such processing is vital in scenarios with moving speakers and/or when very short utterances are processed, e.g., in voice assistant scenarios. We consider several variants of a system that performs beamforming supported by DNN-based voice activity detection (VAD) followed by post-filtering. The speaker is targeted through estimating relative transfer functions between microphones. Each block of the input signals is processed independently in order to make the method applicable in highly dynamic environments. Owing to the short length of the processed block, the statistics required by the beamformer are estimated less precisely. The influence of this inaccuracy is studied and compared to the processing regime when recordings are treated as one block (batch processing). The experimental evaluation of the proposed method is performed on large datasets of CHiME-4 and on another dataset featuring moving target speaker. The experiments are evaluated in terms of objective and perceptual criteria (such as signal-to-interference ratio (SIR) or perceptual evaluation of speech quality (PESQ), respectively). Moreover, word error rate (WER) achieved by a baseline automatic speech recognition system is evaluated, for which the enhancement method serves as a front-end solution. The results indicate that the proposed method is robust with respect to short length of the processed block. Significant improvements in terms of the criteria and WER are observed even for the block length of 250 ms.