Dataset · praharshbhatt
Jarvis-style British English TTS voice checkpoints
praharshbhatt/jarvis-piper-checkpoints
Lightweight Piper (VITS) speech synthesis model checkpoints with a British accent for real-time CPU text-to-speech.
What's inside
This repository contains text-to-speech model checkpoints based on the Piper/VITS architecture (approx. 22M parameters, 63 MB in ONNX format). Distilled from a larger F5-TTS model, it supports British English (eSpeak en-gb-x-rp) and generates mono audio at 22,050 Hz.
What you can do with it
- Integrate a fast English voice engine for a CPU-based voice assistant
- Fine-tune or further distill compact VITS speech synthesis models
- Tasks
- Text → speech
- Languages
- English
- Size
- up to 1K rows
- Downloads this month
- 1K
- Likes
- 3
How to load it
from datasets import load_dataset
ds = load_dataset("praharshbhatt/jarvis-piper-checkpoints")