Automatic Speech Recognition
Safetensors
VibeVoice
vibevoice.c
qwen3
speculative-decoding
dflash
drafter
speech-recognition
awq
compressed-tensors
Instructions to use Ar4ikov/VibeVoice-ASR-DFlash2-Drafter-AWQ-W4A16-ASYM with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- VibeVoice
How to use Ar4ikov/VibeVoice-ASR-DFlash2-Drafter-AWQ-W4A16-ASYM 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("Ar4ikov/VibeVoice-ASR-DFlash2-Drafter-AWQ-W4A16-ASYM") model = VibeVoiceForConditionalGenerationInference.from_pretrained( "Ar4ikov/VibeVoice-ASR-DFlash2-Drafter-AWQ-W4A16-ASYM", 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 config.json from Ar4ikov/VibeVoice-ASR-DFlash2-Drafter-AWQ-W4A16-ASYM: direct link, hf CLI and curl.
- Browser
- Download file 2.19 kB
-
https://huggingface.co/Ar4ikov/VibeVoice-ASR-DFlash2-Drafter-AWQ-W4A16-ASYM/resolve/main/config.json
- Command line
-
hf download hf://Ar4ikov/VibeVoice-ASR-DFlash2-Drafter-AWQ-W4A16-ASYM/config.json
-
curl -L -o config.json https://huggingface.co/Ar4ikov/VibeVoice-ASR-DFlash2-Drafter-AWQ-W4A16-ASYM/resolve/main/config.json
2.19 kB
| { | |
| "architectures": [ | |
| "DFlash2DraftModel" | |
| ], | |
| "model_type": "qwen3", | |
| "attention_bias": false, | |
| "hidden_act": "silu", | |
| "hidden_size": 3584, | |
| "intermediate_size": 9472, | |
| "num_hidden_layers": 5, | |
| "num_attention_heads": 28, | |
| "num_key_value_heads": 4, | |
| "head_dim": 128, | |
| "rms_norm_eps": 1e-06, | |
| "rope_theta": 1000000.0, | |
| "max_position_embeddings": 131072, | |
| "vocab_size": 152064, | |
| "num_target_layers": 28, | |
| "is_causal": false, | |
| "layer_types": [ | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention", | |
| "full_attention" | |
| ], | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "dtype": "bfloat16", | |
| "dflash_config": { | |
| "block_size": 8, | |
| "mask_token_id": 151662, | |
| "target_layer_ids": [ | |
| 1, | |
| 7, | |
| 13, | |
| 19, | |
| 25 | |
| ], | |
| "conv_kernel_size": 2, | |
| "conv_group_size": 16, | |
| "selector_rank": 256, | |
| "selector_top_k": 16, | |
| "loss_decay_gamma": 4.0, | |
| "draft_vocab_size": 32768 | |
| }, | |
| "quantization_config": { | |
| "quant_method": "compressed-tensors", | |
| "format": "pack-quantized", | |
| "quantization_status": "compressed", | |
| "config_groups": { | |
| "group_0": { | |
| "targets": [ | |
| "Linear" | |
| ], | |
| "weights": { | |
| "num_bits": 4, | |
| "type": "int", | |
| "symmetric": false, | |
| "strategy": "group", | |
| "group_size": 128, | |
| "dynamic": false, | |
| "observer": "minmax" | |
| }, | |
| "input_activations": null, | |
| "output_activations": null | |
| } | |
| }, | |
| "ignore": [ | |
| "candidate_selector.hidden_projection" | |
| ], | |
| "kv_cache_scheme": null | |
| }, | |
| "vv_awq": { | |
| "calibration_traces": 48, | |
| "blocks_per_trace": 64, | |
| "folds": [ | |
| "post_attention_layernorm->gate,up", | |
| "up->down", | |
| "v->o" | |
| ], | |
| "clip_ratios": [ | |
| 1.0, | |
| 0.95, | |
| 0.9, | |
| 0.85, | |
| 0.8, | |
| 0.75, | |
| 0.7, | |
| 0.6499999999999999, | |
| 0.6, | |
| 0.55 | |
| ], | |
| "eval": { | |
| "bf16": { | |
| "base_acc": 5.037244695255553, | |
| "sel_acc": 5.271926304412738 | |
| }, | |
| "awq_int4": { | |
| "base_acc": 5.029535440669138, | |
| "sel_acc": 5.267721268765011 | |
| } | |
| } | |
| } | |
| } |