---
title: VoXtream2
emoji: ⚡
colorFrom: blue
colorTo: green
sdk: gradio
python_version: 3.10
sdk_version: 4.44.1
app_file: app.py
pinned: true
license: cc-by-4.0
short_description: Zero-shot TTS with Controllable Speaking-rate
models:
  - herimor/voxtream2
tags:
  - text-to-speech
  - speech-synthesis
preload_from_hub:
  - herimor/voxtream2 model.safetensors,config.json,phoneme_to_token.json
  - kyutai/moshiko-pytorch-bf16 tokenizer-e351c8d8-checkpoint125.safetensors
  - sarulab-speech/sidon-v0.1 decoder_cuda.pt
  - sarulab-speech/sidon-v0.1 feature_extractor_cuda.pt
---