Главная
Study mode:
on
1
Introduction
2
Wave to Back 20
3
XLSR
4
Comparison
5
Data Distribution
6
Evaluation Results
7
Future Setting
8
Multilingual Translation
9
Multilingual Speech Translation
10
Speech Classification
11
Summary
12
XLSR Models
13
Additional Resources
14
Questions
Description:
Explore XLS-R, a groundbreaking large-scale model for cross-lingual speech representation learning based on wav2vec 2.0, in this informative 27-minute talk by Changhan Wang from Meta AI Research. Delve into the model's impressive capabilities, trained on nearly 500,000 hours of public speech audio across 128 languages. Discover how XLS-R significantly improves state-of-the-art performance in speech translation, recognition, and language identification tasks. Learn about its architecture, data distribution, and evaluation results across various benchmarks. Gain insights into future applications, including multilingual translation and speech classification. Understand the potential impact of XLS-R on advancing speech processing for a wide range of global languages and how the open-source community can contribute to its development.

XLS-R: Large-Scale Cross-Lingual Speech Representation Learning on 128 Languages

HuggingFace
Add to list