Summary of Simulseamless: Fbk at Iwslt 2024 Simultaneous Speech Translation, by Sara Papi and Marco Gaido and Matteo Negri and Luisa Bentivogli
SimulSeamless: FBK at IWSLT 2024 Simultaneous Speech Translation
by Sara Papi, Marco Gaido, Matteo Negri, Luisa Bentivogli
First submitted to arxiv on: 20 Jun 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 This paper presents the FBK’s participation in the Simultaneous Translation Evaluation Campaign at IWSLT 2024, specifically focusing on speech-to-text translation (ST) using SimulSeamless. The model combines AlignAtt and SeamlessM4T in its medium configuration, utilizing the latter as an “off-the-shelf” solution and enabling simultaneous inference through cross-attention-based policy. FBK participated in all Shared Task languages, achieving acceptable or better results compared to previous submissions. Notably, SimulSeamless is released, covering over 143 source languages and 200 target languages. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about a team’s attempt to improve simultaneous translation between languages. They created a new system called SimulSeamless that can translate speech from one language into another in real-time. This system combines two existing models to make it work better. The team tested their system on several languages and did well compared to previous years. What’s cool is that they’re releasing their system so others can use it too! |
Keywords
» Artificial intelligence » Cross attention » Inference » Translation
