Low-latency Speech Enhancement via Speech Token Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xue, Huaying, Peng, Xiulian, Lu, Yan
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913204757594112
author Xue, Huaying
Peng, Xiulian
Lu, Yan
author_facet Xue, Huaying
Peng, Xiulian
Lu, Yan
contents Existing deep learning based speech enhancement mainly employ a data-driven approach, which leverage large amounts of data with a variety of noise types to achieve noise removal from noisy signal. However, the high dependence on the data limits its generalization on the unseen complex noises in real-life environment. In this paper, we focus on the low-latency scenario and regard speech enhancement as a speech generation problem conditioned on the noisy signal, where we generate clean speech instead of identifying and removing noises. Specifically, we propose a conditional generative framework for speech enhancement, which models clean speech by acoustic codes of a neural speech codec and generates the speech codes conditioned on past noisy frames in an auto-regressive way. Moreover, we propose an explicit-alignment approach to align noisy frames with the generated speech tokens to improve the robustness and scalability to different input lengths. Different from other methods that leverage multiple stages to generate speech codes, we leverage a single-stage speech generation approach based on the TF-Codec neural codec to achieve high speech quality with low latency. Extensive results on both synthetic and real-recorded test set show its superiority over data-driven approaches in terms of noise robustness and temporal speech coherence.
format Preprint
id arxiv_https___arxiv_org_abs_2310_08981
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Low-latency Speech Enhancement via Speech Token Generation
Xue, Huaying
Peng, Xiulian
Lu, Yan
Sound
Multimedia
Audio and Speech Processing
Existing deep learning based speech enhancement mainly employ a data-driven approach, which leverage large amounts of data with a variety of noise types to achieve noise removal from noisy signal. However, the high dependence on the data limits its generalization on the unseen complex noises in real-life environment. In this paper, we focus on the low-latency scenario and regard speech enhancement as a speech generation problem conditioned on the noisy signal, where we generate clean speech instead of identifying and removing noises. Specifically, we propose a conditional generative framework for speech enhancement, which models clean speech by acoustic codes of a neural speech codec and generates the speech codes conditioned on past noisy frames in an auto-regressive way. Moreover, we propose an explicit-alignment approach to align noisy frames with the generated speech tokens to improve the robustness and scalability to different input lengths. Different from other methods that leverage multiple stages to generate speech codes, we leverage a single-stage speech generation approach based on the TF-Codec neural codec to achieve high speech quality with low latency. Extensive results on both synthetic and real-recorded test set show its superiority over data-driven approaches in terms of noise robustness and temporal speech coherence.
title Low-latency Speech Enhancement via Speech Token Generation
topic Sound
Multimedia
Audio and Speech Processing
url https://arxiv.org/abs/2310.08981