Skip to content
View manash997's full-sized avatar
๐ŸŽฏ
Focusing
๐ŸŽฏ
Focusing

Block or report manash997

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
manash997/README.md

Hi there ๐Ÿ‘‹

My name is Manash, and I'm a machine learning engineer with a strong focus on NLP and Text to Speech. Currently, I'm actively working on TTS and NLP for Indian languages. I hold a masters degree in computer engineering and have a deep passion for machine learning. Before my current role as an ML Engineer at Paisabazaar, I worked as a Research Associate (NLP) at IIT BHU under the Bhashini Program, an initiative of MeitY, Government of India.

During this time, I:

  • Developed a Python tool for Universal Semantic Representation (USR) of Hindi sentences, integrating POS tagging, Named Entity Recognition, and Dependency Parsing.
  • Trained a state-of-the-art Hindi POS tagger using the mT5-base model.
  • Fine-tuned mT5 for predicting missing vibhaktis and detecting complex predicates in Hindi sentences.
  • Designed a sequence-to-sequence framework using mT5-large to reconstruct Hindi sentences from their linearized USR graph representations.

Pinned Loading

  1. generate-usr generate-usr Public

    Python 2 5

  2. vibhakti_predictor_mt5-base vibhakti_predictor_mt5-base Public

    Python

  3. iasnlp2024 iasnlp2024 Public

    Jupyter Notebook

  4. parler-tts parler-tts Public

    Forked from huggingface/parler-tts

    Inference and training library for high-quality TTS models.

    Python

  5. coqui-ai-TTS coqui-ai-TTS Public

    Forked from idiap/coqui-ai-TTS

    ๐Ÿธ๐Ÿ’ฌ - a deep learning toolkit for Text-to-Speech, battle-tested in research and production

    Python