# Frequency Domain Multi-channel Acoustic Modeling for Distant Speech Recognition

arXiv: https://arxiv.org/abs/1903.05299 · HTML: https://mlsanity.com/paper/1903.05299/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Mar 2019.

## Built on by (most cited first, 8 of 8)

- [Recent Advances in End-to-End Automatic Speech Recognition](https://mlsanity.com/paper/2111.01690/) (2021)
- [End-to-End Multi-Channel Transformer for Speech Recognition](https://mlsanity.com/paper/2102.03951/) (2021)
- [Self-Attention Channel Combinator Frontend for End-to-End Multichannel Far-field Speech Recognition](https://mlsanity.com/paper/2109.04783/) (2021)
- [Dual-Encoder Architecture with Encoder Selection for Joint Close-Talk and Far-Talk Speech Recognition](https://mlsanity.com/paper/2109.08744/) (2021)
- [Do You Listen with One or Two Microphones? A Unified ASR Model for Single and Multi-Channel Audio](https://mlsanity.com/paper/2106.02750/) (2021)
- [ChannelAugment: Improving generalization of multi-channel ASR by training with input channel randomization](https://mlsanity.com/paper/2109.11225/) (2021)
- [Spatial Processing Front-End For Distant ASR Exploiting Self-Attention Channel Combinator](https://mlsanity.com/paper/2203.13919/) (2022)
- [Exploring the Integration of Speech Separation and Recognition with Self-Supervised Learning Representation](https://mlsanity.com/paper/2307.12231/) (2023)

## Abstract (arXiv)

Conventional far-field automatic speech recognition (ASR) systems typically employ microphone array techniques for speech enhancement in order to improve robustness against noise or reverberation. However, such speech enhancement techniques do not always yield ASR accuracy improvement because the optimization criterion for speech enhancement is not directly relevant to the ASR objective. In this work, we develop new acoustic modeling techniques that optimize spatial filtering and long short-term memory (LSTM) layers from multi-channel (MC) input based on an ASR criterion directly. In contrast to conventional methods, we incorporate array processing knowledge into the acoustic model. Moreover, we initialize the network with beamformers' coefficients. We investigate effects of such MC neural networks through ASR experiments on the real-world far-field data where users are interacting with an ASR system in uncontrolled acoustic environments. We show that our MC acoustic model can reduce a word error rate (WER) by~16.5% compared to a single channel ASR system with the traditional log-mel filter bank energy (LFBE) feature on average. Our result also shows that our network with the spatial filtering layer on two-channel input achieves a relative WER reduction of~9.5% compared to conventional beamforming with seven microphones.