Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Buckets new
  • Docs
  • Enterprise
  • Pricing
    • Website
      • Tasks
      • HuggingChat
      • Collections
      • Languages
      • Organizations
    • Community
      • Blog
      • Posts
      • Daily Papers
      • Learn
      • Discord
      • Forum
      • GitHub
    • Solutions
      • Team & Enterprise
      • Hugging Face PRO
      • Enterprise Support
      • Inference Providers
      • Inference Endpoints
      • Storage Buckets

  • Log In
  • Sign Up

openbmb
/
VoxCPM2

Text-to-Speech
VoxCPM
Safetensors
tts
multilingual
voice-cloning
voice-design
diffusion
audio
Model card Files Files and versions
xet
Community
14

Instructions to use openbmb/VoxCPM2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

  • Libraries
  • VoxCPM

    How to use openbmb/VoxCPM2 with VoxCPM:

    import soundfile as sf
    from voxcpm import VoxCPM
    
    model = VoxCPM.from_pretrained("openbmb/VoxCPM2")
    
    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
New discussion
Resources
  • PR & discussions documentation
  • Code of Conduct
  • Hub documentation

cross-lingual transfer without an accent

#14 opened 2 days ago by
Markobes

Why?

4
#13 opened 9 days ago by
eugenee00

Report

#12 opened 26 days ago by
fer32

Language Support

๐Ÿ‘€ 1
#11 opened 27 days ago by
Thangamani

้Ÿณ่‰ฒ่ฎพ่ฎกไบง็”Ÿ็š„้Ÿณ้ข‘ๅ†ๆฌก็”จไบŽๅคๅˆปๆ—ถ็š„้—ฎ้ข˜

1
#10 opened 29 days ago by
eldzwdkl

ONNX exports?

#9 opened 30 days ago by
seedblocks

Thank you to everyone who developed this model.

โค๏ธ 6
#7 opened about 1 month ago by
abbas78

Decesion

2
#6 opened about 1 month ago by
Jezecdavins

16-bit quantization

1
#4 opened about 1 month ago by
semevas

Speed issue in results.

2
#3 opened about 1 month ago by
Hanswalter

8-bit quantization

๐Ÿš€๐Ÿ”ฅ 4
#2 opened about 1 month ago by
amarosnithe

ComfyUI Ready ๐ŸŽ‰

๐Ÿš€๐Ÿ‘ 10
1
#1 opened about 1 month ago by
drbaph
Company
TOS Privacy About Careers
Website
Models Datasets Spaces Pricing Docs