Advances in Multimodal Speech Recognition and Social Media Analysis

The field of natural language processing and social media analysis is moving towards a more comprehensive understanding of linguistic and sociological dynamics. Recent developments have focused on advancing multimodal speech recognition, particularly in low-resource languages, and improving the analysis of social media discourse to understand the dissemination of extremist ideologies. Noteworthy papers include: The introduction of LRW-Persian, a large-scale lipreading dataset for the Persian language, which enables rigorous benchmarking and supports cross-lingual transfer. The creation of the Arabic Little STT dataset, which highlights the need for dedicated child speech benchmarks and inclusive training data in automatic speech recognition development.

Sources

The "Right" Discourse on Migration: Analysing Migration-Related Tweets in Right and Far-Right Political Movements

Multilingual Target-Stance Extraction

LRW-Persian: Lip-reading in the Wild Dataset for Persian Language

Arabic Little STT: Arabic Children Speech Recognition Dataset

LASTIST: LArge-Scale Target-Independent STance dataset

Built with on top of