Make Using Tortoise TTS Faster with Fine-Tuned Models

I've been fine tuning a bunch of Tortoise TTS voices, and found that after a model is trained to a voice, you can lower the sample rate and adjust generation parameters to get equivalent or better sounding output than even the Ultra Fast preset.
Experimental fine tuning in other European languages: • Train Tortoise TTS in ...
Timestamps:
Talking about samples: • Make Using Tortoise TT...
Fine-Tuning: • Make Using Tortoise TT...
Generating: • Make Using Tortoise TT...
Tortoise TTS video: • Voice Cloning with Tor...
AI Lip Sync video: • AI Voice Swap and Lip ...
DeepFilter: github.com/Rik...
MakeMKV: www.makemkv.com/
Ultimate Vocal Remover (demucs and other models for Windows): github.com/Anj...
WaveShop (fast .wav editor for Windows): waveshop.sourc...
Download one of my fine-tuned Tortoise TTS models here:
Base model here: huggingface.co...
Requires custom tokenizer file, pt-t.json (put in ./models/tokenizers, switch tokenizer in settings menu)