Instructions to use openbmb/VoxCPM1.5 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- VoxCPM
How to use openbmb/VoxCPM1.5 with VoxCPM:
import soundfile as sf from voxcpm import VoxCPM model = VoxCPM.from_pretrained("openbmb/VoxCPM1.5") wav = model.generate( text="VoxCPM is an innovative end-to-end TTS model from ModelBest, designed to generate highly expressive speech.", prompt_wav_path=None, # optional: path to a prompt speech for voice cloning prompt_text=None, # optional: reference text cfg_value=2.0, # LM guidance on LocDiT, higher for better adherence to the prompt, but maybe worse inference_timesteps=10, # LocDiT inference timesteps, higher for better result, lower for fast speed normalize=True, # enable external TN tool denoise=True, # enable external Denoise tool retry_badcase=True, # enable retrying mode for some bad cases (unstoppable) retry_badcase_max_times=3, # maximum retrying times retry_badcase_ratio_threshold=6.0, # maximum length restriction for bad case detection (simple but effective), it could be adjusted for slow pace speech ) sf.write("output.wav", wav, 16000) print("saved: output.wav") - Notebooks
- Google Colab
- Kaggle
Missing pytorch_model.bin
The pytorch_model.bin isnt automatically downloaded and doesnt appear to be in this repo. For voxcpm 0.5 pytorch_model.bin is in the repo and is automatically downloaded.
Thanks for reaching out about this! In VoxCPM1.5, we have transitioned from using pytorch_model.bin to the safetensors format for model parameters. This change improves security, loading efficiency, and compatibility in some distributed scenarios.
To use VoxCPM1.5, you’ll need to update your voxcpm code—whether you installed it via GitHub or pip. The updated version supports loading models in the .safetensors format while remaining backward compatible with VoxCPM-0.5B (which still uses pytorch_model.bin).
You can update via:
GitHub: Pull the latest changes from the repo.
pip: Run pip install --upgrade voxcpm.
Once updated, you should be able to load both VoxCPM1.5 and VoxCPM-0.5B without issues.
Feel free to try it out and let us know if you encounter any further problems. We appreciate your feedback!
I just pulled the latest changes in github and i got an error saying it was missing pytorch_model.bin.