Text-to-Speech
Transformers
Safetensors
MLX
English
vibevoice_streaming
Realtime TTS
Streaming text input
Long-form speech generation
8-bit precision
Instructions to use mlx-community/VibeVoice-Realtime-0.5B-8bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mlx-community/VibeVoice-Realtime-0.5B-8bit with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-to-speech", model="mlx-community/VibeVoice-Realtime-0.5B-8bit")# Load model directly from transformers import VibeVoiceStreamingForConditionalGenerationInference model = VibeVoiceStreamingForConditionalGenerationInference.from_pretrained("mlx-community/VibeVoice-Realtime-0.5B-8bit", device_map="auto") - MLX
How to use mlx-community/VibeVoice-Realtime-0.5B-8bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir VibeVoice-Realtime-0.5B-8bit mlx-community/VibeVoice-Realtime-0.5B-8bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
| { | |
| "processor_class": "VibeVoiceStreamingProcessor", | |
| "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/Qwen2.5-0.5B" | |
| } |