Skip to content
View bezokurepo's full-sized avatar
1 follower · 9 following

Block or report bezokurepo

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
bezokurepo/README.md
  • 👋 Hi, I’m @bezokurepo
  • 👀 We design and build language technology for low resource and indigenous languages, with the data and models for the first 50 languages covering up to 2.85 billion speakers.
  • 🌱 Our platform builds / extends Universal Dependencies annotated data pipelines using UDPipe2 and conllu editor. We parse connlu annotated corpus into bi-LSTM models (built with PyTorch) that run on Gaudi2 GPU. We train local language models initally for POS tagging and NER. Models for Sentiment Analysis, Grammar Checking, Spell Checking, Question and Answer, and Search are work in progress. We deploy seamlessly to Cloud and Mobile (Android) using Intel openVINO
  • 💞️ We joined the Intel Liftoff program in early 2024 and joined Denvr Ai Ascend in August 2025. All public data and models are open sourced, see license.txt for details in each section.
  • 📫 You can easily search @bezokurepo for open source models and data in the github search bar. If you need support, email [email protected]

Pinned Loading

  1. language-list language-list Public

  2. models models Public

    All released public models from bezoku labs

  3. data data Public

    Jupyter Notebook

  4. papers papers Public

    Main academic papers and research

  5. conferences conferences Public

    The conferences that have made a big impact on the development of Bezoku