Awesome

Portrait Synthesis From Speech

This repository contains the main code for portrait synthesis from speech experiments associated with Retratista application.

Dependencies

Install python3 and python3-pip.
Install requirements.txt using PyPi :
```
pip3 install -r requirements.txt
```

Alternatively, Dockerfile is provided with all required dependencies :

# build image from dockerfile
docker build . -t portrait_syn_from_speech:latest
# run container from image in interactive session
docker run --runtime=nvidia -it portrait_syn_from_speech:latest /bin/bash

Refer to individual README.md inside each component folder for further installation notes.

System Architecture

<div align=center><img width="100%" height="100%" src="architecture.png"/></div> <div align="center"> Figure(1): Complete system architecture diagram showing the flow between different modules. </div><br>

Refer to the wiki for more information about the project application, architecture and workflow.