Name: Himanshu Maurya
Type: User
Company: University of Edinburgh
Bio: AIML Language Engineer @Apple, MSc SLP at University of Edinburgh, Ex-Lead Machine Learning Engineer @iSchoolConnect
Twitter: lordzuko
Location: Hyderabad, India
Blog: https://in.linkedin.com/in/lordzuko
Himanshu Maurya's Projects
Project files from Corise Recsys at Scale course
Improving the Appropriateness in Cross-Text Prosody Transfer using Human Supervision
Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.
Deep learning for audio processing
PyTorch tutorials and best practices.
An implementation of Microsoft's "FastSpeech 2: Fast and High-Quality End-to-End Text to Speech"
Implementation of FastSpeech2 in JAX
FastAPI server for synthesizing speech using FastSpeech2
:zap: Dynamically generated stats for your github readmes
HiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis
A multilingual intent classficiation service
IVORY is a easy to use Hadoop Cluster setup and Administration Portal
Inference code for LLaMA models
Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.
Create characters in Unity with LLMs!
lordzuko's personal website and blog
Building my second brain
This repo is meant to serve as a guide for Machine Learning/AI technical interviews.
Code for various probabilistic deep learning models
Speech samples to be used as part of qualtrics survey for speech for perception alignment of speaking style project
Aligning latent space of speaking style with human perception using a re-embedding strategy
A streamlit based UI for editing speech
A small neural network to perform language modelling on an example piece of text
A template I use frequently for creating tornado application for serving models