AMOOLYA G
@amoolyag
Data Scientist - Speech Specialist at Keypoint Technologies
Bengaluru, India
Amoolya G is a dedicated Data Scientist and M.Tech graduate in Artificial Intelligence with over 2.5 years of experience in machine learning and speech processing. She specializes in developing end-to-end speech-to-speech models, denoising algorithms, and ASR systems for regional languages. A published author and IEEE Best Paper Award winner, she has a proven track record in research and development within the AI and NLP domains.
Experience
Data Scientist - Speech Specialist
Keypoint Technologies
Worked at Defence Research and Development Organization (DRDO) specializing in speech processing. Led the development of an end-to-end speech-to-speech model. Engineered denoising algorithms like CleanUnet and Deepfilternet to reduce word error rate. Implemented data augmentation techniques and optimized model inference time through quantization and pruning. Developed language and dialect identification modules.
MACHINE LEARNING ENGINEER
Kaizen Secure Voiz Pvt Ltd
Pioneered the development of the first ASR system for the Tulu language using Kaldi's GMM-HMM and DNN-HMM techniques. Developed ASR systems for multiple regional languages including Hindi, Kannada, Tamil, and Telugu. Built models for Lip moving detection using Computer Vision. Constructed Restricted Boltzmann Machines on MNIST and custom audio data. Led a team of four for video data collection and validation.
Education
Presidency University
M.Tech
Artificial Intelligence
The Oxford College of Engineering
B.Tech
Mechatronics Engineering
Licenses & Certifications
Practical Introduction to Machine Learning in python
Udemy
Python for data Science and AI
Coursera
Python
Kaggle