TIGER: Time-frequency Interleaved Gain Extraction and Reconstruction for Efficient Speech Separation

TL;DR

The paper introduces TIGER, an efficient speech separation model that reduces computational costs significantly while outperforming state-of-the-art models, and presents EchoSet, a new dataset for realistic acoustic environment evaluation.

Contribution

TIGER is a novel low-parameter, low-complexity speech separation model utilizing frequency band division and attention modules, validated on a new realistic dataset EchoSet.

Findings

TIGER reduces parameters by 94.3% and MACs by 95.3%.

TIGER outperforms SOTA models on EchoSet and real-world data.

EchoSet provides a more realistic benchmark for speech separation.

Abstract

In recent years, much speech separation research has focused primarily on improving model performance. However, for low-latency speech processing systems, high efficiency is equally important. Therefore, we propose a speech separation model with significantly reduced parameters and computational costs: Time-frequency Interleaved Gain Extraction and Reconstruction network (TIGER). TIGER leverages prior knowledge to divide frequency bands and compresses frequency information. We employ a multi-scale selective attention module to extract contextual features while introducing a full-frequency-frame attention module to capture both temporal and frequency contextual information. Additionally, to more realistically evaluate the performance of speech separation models in complex acoustic environments, we introduce a dataset called EchoSet. This dataset includes noise and more realistic reverberation (e.g., considering object occlusions and material properties), with speech from two speakers overlapping at random proportions. Experimental results showed that models trained on EchoSet had better generalization ability than those trained on other datasets compared to the data collected in the physical world, which validated the practical value of the EchoSet. On EchoSet and real-world data, TIGER significantly reduces the number of parameters by 94.3% and the MACs by 95.3% while achieving performance surpassing the state-of-the-art (SOTA) model TF-GridNet.