Instructions to use RayNene/VibeVoice-Large with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- VibeVoice
How to use RayNene/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("RayNene/VibeVoice-Large") model = VibeVoiceForConditionalGenerationInference.from_pretrained( "RayNene/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
Download preprocessor_config.json from RayNene/VibeVoice-Large: direct link, hf CLI and curl.
- Browser
- Download file 349 Bytes
-
https://huggingface.co/RayNene/VibeVoice-Large/resolve/main/preprocessor_config.json
- Command line
-
hf download hf://RayNene/VibeVoice-Large/preprocessor_config.json
-
curl -L -o preprocessor_config.json https://huggingface.co/RayNene/VibeVoice-Large/resolve/main/preprocessor_config.json
349 Bytes
| { | |
| "processor_class": "VibeVoiceProcessor", | |
| "speech_tok_compress_ratio": 3200, | |
| "db_normalize": true, | |
| "audio_processor": { | |
| "feature_extractor_type": "VibeVoiceTokenizerProcessor", | |
| "sampling_rate": 24000, | |
| "normalize_audio": true, | |
| "target_dB_FS": -25, | |
| "eps": 1e-06 | |
| }, | |
| "language_model_pretrained_name": "Qwen/Qwen2.5-7B" | |
| } |