Usamireko

Yet another human from the Outside World

Skills & Expertise

AI & Machine Learning

  • Neural Networks
  • Singing Voice Synthesis
  • Text-to-Speech
  • Generative AI

Development

  • Python
  • Basic Web Development
  • Usage of OpenAI-like APIs
  • C#

AI Showcase

StableTTS Training

A somewhat comprehensive guide and implementation for training StableTTS models on Google Colab.

View Project

HiFiPLN Colab

Google Colab notebook for HiFiPLN vocoder fine-tuning.

View Project

DiffusionDenoiser

Audio denoising using Fish Diffusion.

View Project

Non-English Tacotron-2

Training notebook for Tacotron-2 models with non-English language support.

View Project

DiffSinger Colab

Fork of the MLo7´s DiffSinger Colab notebook with some modifications.

View Project

YuE-exllamav2

Optimized implementation of multimodal-art-projection/YuE in exllamav2 now with a Colab notebook.

View Project

Contact

Feel free to reach out for collaborations or just to say hello!

Say Hello!
Usamireko's Profile Picture

About Me

AI & Vocal Synthesis Enthusiast

Basic Information

Name Usamireko/Sumi/Lin
Pronouns Shhh, lets keep it as a secret c:
Location Outside World
Occupation University Student

Technical Background

Main Focus AI & Vocal Synthesis
Languages English, Spanish
Development Python, C#, Web

Interests & Hobbies

Artificial Intelligence
Singing Voice Synthesis & Text-to-Speech
Open Source Development
Touhou Project

Bio

I'm a university student passionate about mixing technology and creativity. My vocal synthesis journey began as a kid when discovering Hatsune Miku, when AI and I crossed paths, I knew that I wanted to work with it as an adult. I love exploring the limits of what's possible with current technology and contributing to the community.