Home Research Papers Datasets State of the Art Pricing

Discover, visualize, and connect AI research papers. Explore the latest trends and insights in artificial intelligence research.

Product

Home
Research Papers
About

Support

Contact
Terms of Service
Privacy Policy

speech-1

Speech Extraction

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in speech-extraction-1

Trend

Dataset

Best Model

Actions

WSJ0-2mix-extr

Libraries

Use these libraries to find speech-extraction-1 models and implementations

Datasets

WSJ0-2mix-extr

MC_GRID

WHAMR_ext

Subtasks

No subtasks available.

Most implemented papers

Improving Speaker Discrimination of Target Speech Extraction With Time-Domain Speakerbeam

Marc Delcroix, Tsubasa Ochiai, Kateřina Žmolíková, K. Kinoshita, Naohiro Tawara, T. Nakatani, S. Araki•Wed Jan 22 2020

Strategies for improving the speaker discrimination capability of SpeakerBeam are investigated and it is shown experimentally that these strategies greatly improve speech extraction performance, especially for same-gender mixtures, and outperform TasNet in terms of target speech extraction.

145

Content

Introduction Benchmarks Datasets Subtasks Libraries Papers

Paper Graph

Multi-Channel Target Speech Extraction with Channel Decorrelation and Target Speaker Adaptation

Jiangyu Han, Xinyuan Zhou, Yanhua Long, Yijie Li•Sun Oct 18 2020

This work proposes two methods for exploiting the multi-channel spatial information to extract the target speech by using a target speech adaptation layer in a parallel encoder architecture and designing a channel decorrelation mechanism to Extract the inter-channel differential information to enhance theMulti-channel encoder representation.

15 0

Paper Graph

Target Speech Extraction Based on Blind Source Separation and X-vector-based Speaker Selection Trained with Data Augmentation.

Kai-Jyun Chen, Jing Lu, Lele Liao, Zhaoyi Gu•Fri May 15 2020

This paper explores a sequential approach for target speech extraction by combining blind source separation (BSS) with the x-vector based speaker recognition (SR) module, and extends the training of MVAE to evaluate its generalization to unseen speakers.

3 0

Paper Graph

DPCCN: Densely-Connected Pyramid Complex Convolutional Network for Robust Speech Separation and Extraction

J. Černocký, L. Burget, Jiangyu Han, Yanhua Long•Sun Dec 26 2021

A densely-connected pyramid complex convolutional network, termed DPCCN, is proposed to improve the robustness of speech separation under complicated conditions and is generalized to target speech extraction (TSE) by integrating a new specially designed speaker encoder.

33 0

Paper Graph

On the Role of Spatial, Spectral, and Temporal Processing for DNN-based Non-linear Multi-channel Speech Enhancement

Timo Gerkmann, Kristina Tesch, Nils-Hendrik Mohrmann•Tue Jun 21 2022

The experiments in a difficult speech extraction scenario confirm the importance of non-linear spatial filtering, which outperforms an oracle linear spatial filter by 0.24 POLQA score, and demonstrate that joint processing results in a large performance gap of 0.4 POLZA score between network architectures exploiting spectral versus temporal information besides spatial information.

10 0

Paper Graph

Analysis of Impact of Emotions on Target Speech Extraction and Speech Separation

M. Kocour, Marc Delcroix, Tsubasa Ochiai, Jan vCernock'y, Ladislav Movsner, Ján Švec, Katevrina vZmol'ikov'a•Sun Aug 14 2022

It is observed that BSS is relatively robust to emotions, while TSE, which requires identifying and extracting the speech of a tar-get speaker, is much more sensitive to emotions.

6 0

Paper Graph

Neural Target Speech Extraction: An overview

Dong Yu, Marc Delcroix, Tsubasa Ochiai, Kateřina Žmolíková, K. Kinoshita, Jan vCernock'y•Mon Jan 30 2023

This article focuses on recent neural-based approaches in target speech/speaker extraction and presents an in-depth overview of TSE, guiding readers through the different major approaches, emphasizing the similarities among frameworks and discussing potential future directions.

137 0

Paper Graph

Look Once to Hear: Target Speech Hearing with Noisy Examples

Bandhav Veluri, Tuochao Chen, Shyamnath Gollakota, Malek Itani, Takuya Yoshioka•Thu May 09 2024

This paper presents the first enrollment interface where the wearer looks at the target speaker for a few seconds to capture a single, short, highly noisy, binaural example of the target speaker, enabling target speech hearing to ignore all interfering speech and noise, but the target speaker.

32 0

Paper Graph

Adding a benchmark result helps the community track progress.

Speech Extraction | State-of-the-Art