# Dual-Path Filter Network: Speaker-Aware Modeling for Speech Separation

arXiv: https://arxiv.org/abs/2106.07579 · HTML: https://mlsanity.com/paper/2106.07579/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Jun 2021.

## Built on by (most cited first, 4 of 4)

- [X-SepFormer: End-to-end Speaker Extraction Network with Explicit Optimization on Speaker Confusion](https://mlsanity.com/paper/2303.05023/) (2023)
- [Directed Speech Separation for Automatic Speech Recognition of Long Form Conversational Speech](https://mlsanity.com/paper/2112.05863/) (2021)
- [Individualized Conditioning and Negative Distances for Speaker Separation](https://mlsanity.com/paper/2210.06368/) (2022)
- [Improving Target Speaker Extraction with Sparse LDA-transformed Speaker Embeddings](https://mlsanity.com/paper/2301.06277/) (2023)

## Abstract (arXiv)

Speech separation has been extensively studied to deal with the cocktail party problem in recent years. All related approaches can be divided into two categories: time-frequency domain methods and time domain methods. In addition, some methods try to generate speaker vectors to support source separation. In this study, we propose a new model called dual-path filter network (DPFN). Our model focuses on the post-processing of speech separation to improve speech separation performance. DPFN is composed of two parts: the speaker module and the separation module. First, the speaker module infers the identities of the speakers. Then, the separation module uses the speakers' information to extract the voices of individual speakers from the mixture. DPFN constructed based on DPRNN-TasNet is not only superior to DPRNN-TasNet, but also avoids the problem of permutation-invariant training (PIT).