Algorithm Intern (Character Modeling) – MSRA Speech Group

Date Posted: October 14, 2019

Roles & Responsibilities

·        Work on the scene text recognition project

·        Investigate on the end2end character recognition approaches and train the character models

Required Qualifications:

·        Grad student with good knowledge on deep learning, especially on image/OCR/speech recognition

·        Familiar with the end-to-end modeling approaches, e.g., CTC, encoder-attention-decoder and RNN Transducer

·         Good python programing skills

·        Team work and good communication skills


·        Have experiences in training models with large scale training data

·        6+ month full time internship