P
VieNeu-TTS v2 Turbo
bởi pnnbao-umpCông khaiapache-2.0
README.md
VieNeu-TTS v2 Turbo
VieNeu-TTS v2 Turbo is an advanced, lightweight bilingual Vietnamese-English text-to-speech and voice cloning model, published by pnnbao-ump on Hugging Face.
Links
- Upstream model card: https://huggingface.co/pnnbao-ump/VieNeu-TTS-v2
- Runtime GitHub: https://github.com/dduongtrandai/VieNeu-TTS.cpp
Model Facts
- Task: Text-to-Speech & Voice Cloning
- Parameters: ~300M
- License: Apache-2.0
- Architecture: VieNeu-TTS (GGUF)
LA Studio Notes
VieNeu-TTS v2 Turbo in LA Studio is optimized for fast CPU execution and instant 3-5 seconds zero-shot voice cloning. It downloads files from conversion repositories:
pnnbao-ump/VieNeu-TTS-v2-Turbo-GGUFfor the GGUF backbone and preset voices configuration.pnnbao-ump/VieNeu-Codecfor the speaker encoder and neural decoder ONNX models.
Cấu hình tham số quy trình
Cấu hình các tham số tùy chỉnh này để tinh chỉnh quy trình tiếng nói và giọng nói.
Temperaturefloat
Controls randomness of speech generation. Higher values increase variety.
1
Seedint
Random seed for reproducibility. Set to -1 for random.
-1
Max Codec Stepsint
Maximum decoding steps for the neural voice audio synthesizer.
100