I enjoy making things. Here are a selection of projects that I have worked on over the years.
for finetuning pre-trained vocos model with mels genarated from any tts decoder use
Speech to text, text to intent. REST API application that can be deployed in cloud.
This is an experiment to check if we can clone a voice for the VITS tts. Here we will use tts models from MMS.
A speech synthesis system with prosody embeddings.
