Summary of Xls-r Deep Learning Model For Multilingual Asr on Low- Resource Languages: Indonesian, Javanese, and Sundanese, by Panji Arisaputra et al.
XLS-R Deep Learning Model for Multilingual ASR on Low- Resource Languages: Indonesian, Javanese, and Sundanese
by Panji Arisaputra, Alif Tri Handoyo, Amalia Zahra
First submitted to arxiv on: 12 Jan 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Machine Learning (cs.LG); 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 The study develops and evaluates an Automatic Speech Recognition (ASR) system, specifically designed for Indonesian, Javanese, and Sundanese languages. Using the XLS-R 300m model, the researchers aimed to improve ASR performance in converting spoken language into written text. The methodology employed includes testing procedures, datasets used, and training techniques. The results show competitive Word Error Rate (WER) measurements, with some compromise in performance for Javanese and Sundanese languages. Integrating a 5-gram KenLM language model significantly reduces WER and enhances ASR accuracy. This research contributes to advancing ASR technology by addressing linguistic diversity and improving performance across various languages. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study makes an automatic speech recognition (ASR) system better for Indonesian, Javanese, and Sundanese languages. The researchers use a special model called XLS-R 300m to improve how well the system works. They test it on different languages and find that it does pretty well, but not perfectly. Adding another tool called KenLM makes it even better! This research helps make ASR systems more useful for people who speak these languages. |
Keywords
* Artificial intelligence * Language model
