Downloads · 30 days
0
gpt-sovits-v2pp-onnx
jiangzhuo9357/gpt-sovits-v2pp-onnx
gpt-sovits-v2pp-onnx is a machine learning model from jiangzhuo9357. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Converted from the base pretrained checkpoints of GPT-SoVITS (MIT) — lj1995/GPT-SoVITS (MIT) — via Genie-TTS (MIT). Runtime assets (chinese-hubert-base, speaker encoder, RoBERTa, G2P dictionaries)
Access
Public
Updated Jul 16, 2026
Repo size
1.3 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.onnx787 MB · 58%
At a glance
- Access
- Public
- Created
- Jul 16, 2026
- Updated
- Jul 16, 2026
- SHA
- 516a13cc
Model card
From the Hugging Face model README
GPT-SoVITS v2ProPlus — ONNX (Sokuji Local Native TTS)
Converted from the base pretrained checkpoints of GPT-SoVITS (MIT) — lj1995/GPT-SoVITS (MIT) — via Genie-TTS (MIT). Runtime assets (chinese-hubert-base, speaker encoder, RoBERTa, G2P dictionaries) mirrored from High-Logic/Genie (MIT; hubert originally TencentGameMate/chinese-hubert-base, MIT; RoBERTa Apache-2.0).
Weight bins under model/ and genie_data/chinese-hubert-base/ are fp16;
the Sokuji sidecar expands them to fp32 in place at load time.
Default voice clips: en — JFK 1961 inaugural address excerpt (US government work, public domain); zh/ja — fully synthetic speech generated with Qwen3-TTS 1.7B (Apache-2.0), no human speaker involved.
- Created
- Jul 16, 2026
- Updated
- Jul 16, 2026