Summary of Task Arithmetic For Language Expansion in Speech Translation, by Yao-fei Cheng et al.
Task Arithmetic for Language Expansion in Speech Translation
by Yao-Fei Cheng, Hayato Futami, Yosuke Kashiwagi, Emiru Tsunoo, Wen Shen Teo, Siddhant Arora, Shinji Watanabe
First submitted to arxiv on: 17 Sep 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
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 The proposed approach expands language pairs in instruction-based speech translation by merging models trained on new and existing datasets using task arithmetic. The method eliminates language confusion by introducing an augmented task arithmetic model that generates correct target language tokens following instructions. Experiments demonstrate a BLEU score improvement of up to 4.92 in CoVoST-2 and 4.66 in MuST-C, while also expanding to language pairs without paired ST training data or pre-trained models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A team of researchers found a way to make speech-to-text translation work with new languages without having to train the system all over again. They did this by combining two types of machine learning models: one that knows how to translate, and another that can understand what language is being spoken. This allowed them to create a new system that can translate between many different languages. The new system was tested and shown to be effective in translating speech into text with high accuracy. |
Keywords
» Artificial intelligence » Bleu » Machine learning » Translation
