Separate and Reconstruct: Asymmetric Encoder-Decoder for Speech Separation

Feb-14-2026, 17:47:47 GMT–Neural Information Processing Systems

In speech separation, time-domain approaches have successfully replaced the time-frequency domain with latent sequence feature from a learnable encoder.

artificial intelligence, machine learning, natural language, (21 more...)

Neural Information Processing Systems

Feb-14-2026, 17:47:47 GMT

Conferences PDF

Country:
- Asia
  - China > Hong Kong (0.04)
  - South Korea > Seoul
    - Seoul (0.04)
- Europe
  - France > Île-de-France
    - Paris > Paris (0.04)
  - Germany > Bavaria
    - Upper Bavaria > Munich (0.04)
  - Italy > Tuscany
    - Florence (0.04)

Genre:
- Research Report
  - Experimental Study (0.93)
  - New Finding (0.93)

Technology:
- Information Technology
  - Artificial Intelligence
    - Machine Learning > Neural Networks
      - Deep Learning (0.68)
    - Natural Language (1.00)
    - Speech > Speech Recognition (0.68)
    - Vision (0.68)
  - Sensing and Signal Processing > Image Processing (0.67)

Duplicate Docs Excel Report

Title
Separate and Reconstruct: Asymmetric Encoder-Decoder for Speech Separation

Similar Docs Excel Report more

Title	Similarity	Source
None found