Summary of Continuously Learning New Words in Automatic Speech Recognition, by Christian Huber and Alexander Waibel
Continuously Learning New Words in Automatic Speech Recognition
by Christian Huber, Alexander Waibel
First submitted to arxiv on: 9 Jan 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG)
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 This paper proposes a self-supervised continual learning approach to improve Automatic Speech Recognition (ASR) systems’ ability to recognize acronyms, named entities, and domain-specific special words. By using a memory-enhanced ASR model and leveraging labeled data from slides, the system learns to decode new words and adapt to novel vocabulary. The approach iteratively trains on a growing dataset of utterances containing detected new words, achieving over 80% recall for these words while maintaining general performance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps fix speech recognition problems by teaching machines to learn from mistakes. Right now, speech recognition isn’t perfect because it often misses special words like acronyms and names. To solve this issue, the researchers created a system that uses lecture slides to help the machine learn new words. The system looks at what it got wrong in the past, adds those mistakes to its training data, and then gets better at recognizing similar errors. This approach helps improve speech recognition accuracy for special words. |
Keywords
* Artificial intelligence * Continual learning * Recall * Self supervised
