---
title: NanoTTS
emoji: 🐠
colorFrom: yellow
colorTo: indigo
sdk: gradio
sdk_version: 6.14.0
python_version: '3.10'
app_file: app.py
pinned: false
license: mit
short_description: minimal text-to-speech (TTS) using nanoGPT
suggested_hardware: t4-small
preload_from_hub:
  - novateur/WavTokenizer wavtokenizer_smalldata_frame40_3s_nq1_code4096_dim512_kmeans200_attn.yaml,WavTokenizer_small_600_24k_4096.ckpt
---

# nanoTTS

A Gradio demo for nanoTTS, a minimal GPT-style text-to-speech model that generates WavTokenizer audio tokens from text.