realtime-voice-accent-improvement

An attempt to improve some videos voice audio accent in english language; at realtime

How many times, we found marvelouns information on video, but, BUT, when we start to see that media, KAPUT, the voice audio accent destroys ours ears; that's not meaning I have a great accent, in English Language, maybe for those with those accents, WE HAVE THE WORST... but, I want to AI and NVIDIA Workbench comes to rescue MY DAMAGED EARs, period.

No body was hurt in the making of this project.

Running the TTS Server

I will run this project, using a notebook into the jupyter lab environment provided by Nvidia AI Workbench.

From there, I will call the coqui-ai TTS server, running from docker. How? I show You in details, with images or better, a Video about HOW TO run the TTS Server from docker.

1. docker run --rm -it -p 5002:5002 --entrypoint /bin/bash ghcr.io/coqui-ai/tts-cpu #pull the server (if you don't have it) and then run in port 5002
2. python3 TTS/server/server.py --list_models #To get the list of available models
3. python3 TTS/server/server.py --model_name tts_models/en/vctk/vits # To start a server --mmodel_name tts_models/en/vctk/vits (the name of the model you're going to use from 2.

Install TTS into the jupyter lab NVIDIA AI Workbench

Just run pip install TTS from a notebook, or add packacge from the windows within NIVIDA AI Workbench: Package, Add, select pip, then put TTS