Instructions to use aoi-ot/VibeVoice-Large with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- VibeVoice
How to use aoi-ot/VibeVoice-Large with VibeVoice:
import torch, soundfile as sf, librosa, numpy as np from vibevoice.processor.vibevoice_processor import VibeVoiceProcessor from vibevoice.modular.modeling_vibevoice_inference import VibeVoiceForConditionalGenerationInference # Load voice sample (should be 24kHz mono) voice, sr = sf.read("path/to/voice_sample.wav") if voice.ndim > 1: voice = voice.mean(axis=1) if sr != 24000: voice = librosa.resample(voice, sr, 24000) processor = VibeVoiceProcessor.from_pretrained("aoi-ot/VibeVoice-Large") model = VibeVoiceForConditionalGenerationInference.from_pretrained( "aoi-ot/VibeVoice-Large", torch_dtype=torch.bfloat16 ).to("cuda").eval() model.set_ddpm_inference_steps(5) inputs = processor(text=["Speaker 0: Hello!\nSpeaker 1: Hi there!"], voice_samples=[[voice]], return_tensors="pt") audio = model.generate(**inputs, cfg_scale=1.3, tokenizer=processor.tokenizer).speech_outputs[0] sf.write("output.wav", audio.cpu().numpy().squeeze(), 24000) - Notebooks
- Google Colab
- Kaggle
Update library tag for better download tracking and code snippets!
#7
by Steveeeeeeen HF Staff - opened
Hi, I'm Steven from HF! I opened a small PR to help get clearer and better snippets + library_name
cf: https://huggingface.co/aoi-ot/VibeVoice-7B/discussions/1
https://huggingface.co/aoi-ot/VibeVoice-1.5B/discussions/1
Steveeeeeeen changed pull request title from Update README.md to Update library tag for better download tracking and code snippets!
Note: library_name must be lowercased for it to work.
library_name: vibevoice
@Steveeeeeeen could you update this PR? π
aoi-ot changed pull request status to merged