Summary of Query-by-example Keyword Spotting Using Spectral-temporal Graph Attentive Pooling and Multi-task Learning, by Zhenyu Wang et al.
Query-by-Example Keyword Spotting Using Spectral-Temporal Graph Attentive Pooling and Multi-Task Learning
by Zhenyu Wang, Shuyu Kong, Li Wan, Biqiao Zhang, Yiteng Huang, Mumin Jin, Ming Sun, Xin Lei, Zhaojun Yang
First submitted to arxiv on: 27 Aug 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Sound (cs.SD); Audio and Speech Processing (eess.AS)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel Query-by-Example (QbyE) keyword spotting (KWS) system is proposed, which leverages spectral-temporal graph attentive pooling and multi-task learning to learn speaker-invariant and linguistic-informative embeddings. The framework involves investigating three distinct network architectures for encoder modeling: LiCoNet, Conformer, and ECAPA_TDNN. Experimental results on an internal dataset of 629 speakers demonstrate the effectiveness of the proposed QbyE framework in maximizing the potential of simpler models like LiCoNet. Notably, LiCoNet achieves comparable performance to the computationally intensive Conformer model while being 13x more efficient (1.98% vs. 1.63% FRR at 0.3 FAs/Hr). |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new way is found to make computers recognize custom words spoken by people. This system uses a special kind of learning that helps it understand how words sound, even when different people say them. The researchers tested three different ways of doing this and found one method, called LiCoNet, was very good at recognizing words. It’s like a shortcut that makes the computer work faster without losing accuracy. |
Keywords
» Artificial intelligence » Encoder » Multi task
