PulseAugur
EN
LIVE 22:22:05

ai-sage releases GigaAM Multilingual speech models

ai-sage has released GigaAM Multilingual, a family of Conformer-based foundation models. These models, available in 220M and 600M parameter variants, have been pre-trained on over 2 million hours of speech data spanning more than 70 languages. They are fine-tuned for automatic speech recognition (ASR) and offer top-tier open-source performance for Russian, Kazakh, Kyrgyz, and Uzbek, with moderate performance for English. AI

IMPACT Provides high-quality open-source ASR models for underrepresented languages, potentially accelerating research and development in those regions.

RANK_REASON Release of an open-source speech model with performance details and usage instructions. [lever_c_demoted from research: ic=1 ai=1.0]

Read on Hugging Face Trending Models →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

ai-sage releases GigaAM Multilingual speech models

COVERAGE [1]

  1. Hugging Face Trending Models TIER_1 (TL) · ai-sage ·

    ai-sage/GigaAM-Multilingual

    automatic-speech-recognition · 2,399 downloads · 51 likes