Downloads · 30 days
0
kitten-nano-en-v0_2-fp16
csukuangfj2/kitten-nano-en-v0_2-fp16
kitten-nano-en-v0_2-fp16 is a machine learning model from csukuangfj2. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
See also https://github.com/KittenML/KittenTTS
Access
Public
Updated May 12, 2026
Repo size
38 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.onnx23.8 MB · 57%
At a glance
- Access
- Public
- Created
- May 12, 2026
- Updated
- May 12, 2026
- SHA
- ad1357e6
Model card
From the Hugging Face model README
Introduction
See also https://github.com/KittenML/KittenTTS
KittenTTS v0.8
Use v0_8/run.sh to prepare a sherpa-onnx compatible model directory from
the upstream Hugging Face assets:
cd scripts/kitten-tts/v0_8
python3 -m pip install numpy onnx
./run.sh KittenML/kitten-tts-nano-0.8-fp32
./run.sh KittenML/kitten-tts-nano-0.8-int8
./run.sh KittenML/kitten-tts-micro-0.8
./run.sh KittenML/kitten-tts-mini-0.8
The generated voices.bin preserves all v0.8 reference rows. The ONNX metadata
marks v0.8 models with version=8, end_id=10, add_pad_after_end=1, and
max_token_len=400 so the runtime selects the same style row as upstream.
Use package directories like kitten-mini-en-v0_8,
kitten-micro-en-v0_8, kitten-nano-en-v0_8-fp32, or
kitten-nano-en-v0_8-int8.
- Created
- May 12, 2026
- Updated
- May 12, 2026