The Mozilla Data Collective is hosting a model competition with a $20,000 prize pool focused on Automatic Speech Recognition (ASR). The competition specifically targets the challenge of code-switching, where speakers use multiple languages or dialects within a single utterance. Participants are tasked with developing ASR systems that can accurately transcribe code-switched speech across three language pairs: North American Spanish and English, Indonesian and Javanese, and Spanish and Nahuatl, aiming to advance the state-of-the-art for these under-served linguistic communities. AI
IMPACT This competition aims to improve ASR capabilities for under-served languages, potentially leading to more inclusive speech technology.
RANK_REASON Competition announcement focused on advancing ASR technology for specific linguistic challenges. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Mastodon — sigmoid.social →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
